The colorful giants: Revisiting the systematics of the Anolis latifrons series (Squamata: Anolidae)

441 The colorful giants: Revisiting the systematics of the Anolis latifrons series (Squamata: Anolidae) Carlos M. Marín1, Daniel Bocanumenth2, Juan M. Daza2 1 Laboratorio de Macroecología Evolutiva, Red de Biología Evolutiva, Instituto de Ecología A.C., Veracruz, CP, 91073, Mexico 2 Grupo Herpetológico de Antioquia GHA, Instituto de Biología, Facultad de Ciencias Exactas y Naturales, Universidad de Antioquia UdeA, Calle 70 No. 52-21, Medellín, Colombia https://zoobank.org/8134FAA0-D9C8-4682-94BE-644982083A9F Corresponding author: Juan M. Daza ([email protected]) Academic editor Uwe Fritz | Received 17 June 2025 | Accepted 28 September 2025 | Published 29 October 2025 Citation: Marín CM, Bocanumenth D, Daza JM (2025) The colorful giants: Revisiting the systematics of the Anolis latifrons series (Squamata: Anolidae). Vertebrate Zoology 75: 441–457. https://doi.org/ 10.3897/vz.75.e162071 Abstract The lizard genus Anolis is the second most diverse genus of terrestrial vertebrates. Within Anolis, the highly speciose clade Dactyloa comprises six species series, including the latifrons series. Despite previous efforts to reconstruct its phylogeny, earlier studies have excluded a substantial proportion of the clade’s species diversity. Here, we integrated both historical and newly generated genetic data to reconstruct the most comprehensive molecular phylogeny of the latifrons series to date including 88% of the current species diversity. We also conducted a thorough morphological examination of museum specimens representing ten species, primarily distributed in Colombia, including vouchers of A. danieli from the Central Cordillera used in previous molecular phylogenies. Our phylogeny also included genetic samples of A. danieli from several localities in the Western Cordillera, samples of Anolis limon and A. mirus (two species previously lacking genetic data), and sequences from the Central American species A. kathydayae and A. brooksi. Our results recovered topological differences for A. limon and A. mirus compared to previous hypotheses and revealed that specimens assigned to A. danieli in earlier studies were misidentified and are not phylogenetically related to this species. Instead, our results showed that the true A. danieli is sister to a green anole clade distributed across the Central Cordillera, Pacific region, and Panama. Based on our phylogenetic and genetic distance analyses, we conclude that A. kathydayae should be considered a junior synonym of A. brooksi. Lastly, we describe the taxon previously confused with A. danieli and comment on the taxonomic implications of our findings. Keywords Biogeography, Dactyloa, morphometrics, Neotropics, species description, systematics Introduction The Neotropical lizard genus Anolis Daudin, 1802, with 434 species (Uetz et al. 2025), is recognized as the second most diverse genus of terrestrial vertebrates after the rainfrog genus Pristimantis Jiménez De La Espada, 1870 (Poe et al. 2017; Christodoulides et al. 2024). The outstanding species richness, high levels of cryptic diversity (Yánez-Muñoz et al. 2018; Moreno-Arias et al. 2023) and incomplete geographic and genetic sampling in mainland regions across South America, including the Andes and the Amazonia, have hampered accurate species delimitation within several groups (Velasco and Hurtado-Gómez 2014; Grisales-Martínez et al. 2017; Prates et al. 2020). As a result, a substantial proportion of the genus diversity is still to be discovered. Within Anolis, the clade DacVertebrate Zoology 75, 2025, 441–457 | DOI 10.3897/vz.75.e162071 Copyright Carlos M. Marín. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Marin CM et al.: Anolis Systematics 442 tyloa Wagler, 1830 comprises large-bodied species primarily found on the South American mainland, as well as a smaller, less diverse radiation in the Lesser Antilles (Castañeda and de Queiroz 2011, 2013). Currently, six species groups (Williams 1976) or series (Savage and Guyer 1989) are recognized within Dactyloa, including the punctatus, roquet, heterodermus, aequatorialis, and latifrons series and the more recently proposed nasofrontalis series by Prates et al. (2020). The latifrons series was first proposed by Etheridge (1960) to accommodate three species: Anolis latifrons Berthold, 1846, A. frenatus Cope, 1899 and A. fraseri Günther, 1859. The taxonomic content of this series, as originally circumscribed, remained largely unchanged until molecular and morphological data began to be generated by studies that tested its monophyly (Poe 2004; Velasco and Hoyos 2010; Castañeda and de Queiroz 2011, 2013). The species content of the series has varied significantly as recent studies have added new species to the group (Poe et al. 2015; Poe and Ryan 2017; Prates et al. 2020). Currently, the latifrons series includes 25 named species distributed across the lowlands of Panama, Costa Rica, Venezuela, Ecuador, the Pacific region (including Malpelo island), and the inter-Andean valleys of Colombia (Castañeda and de Queiroz 2011, 2013; Poe et al. 2017; Prates et al. 2020). The most comprehensive phylogeny of the latifrons series to date, published by Poe et al. (2017), assessed evolutionary relationships by incorporating morphological data for species lacking DNA sequences into a primarily molecular framework. This approach raises concerns about the reliability of phylogenetic placements for taxa without genetic data, particularly when evaluated from a strictly molecular standpoint. More importantly, the phylogeny of Poe et al. (2017), along with subsequent studies based exclusively on molecular data (e.g., Batista et al. 2015; Prates et al. 2017, 2020; Ayala-Varela et al. 2021), failed to include a substantial portion of the group’s species diversity. For example, Poe et al. (2017) included 20 of the 24 species recognized at the time; however, DNA sequences were unavailable for 25% of the evaluated taxa. More recent molecular phylogenies have included between 52% and 80% of the species (Prates et al. 2017; Ayala-Varela et al. 2021). Remarkably, the more recent studies omitted A. brooksi (Barbour, 1923) despite the availability of molecular data for this taxon. This situation underscores the need for an updated phylogeny that incorporates most of the currently recognized diversity in the series and evaluates the placement of species previously excluded from molecular phylogenies due to missing sequence data. Despite efforts to reconstruct the phylogeny of the lati frons series (e.g., Velasco and Hoyos 2010; Castañeda and Queiroz 2011, 2013; Prates et al. 2017, 2020) several gaps remain concerning the taxonomic identity, phylogenetic position, and geographical distribution of several species. One such example is Anolis danieli Williams, 1988, originally described from the municipality of Urrao in Antioquia, Colombia. In addition to the type locality, this species is known from several other localities in the northwestern Cordillera in Colombia (Velasco and Hoyos 2010). Nonetheless, despite its well-known distribution, the only genetic samples labeled as Anolis danieli, included for the first time in the phylogenetic analyses by Castañeda and de Queiroz (2011) and then in a series of subsequent studies (see Castañeda and de Queiroz 2013; Poe et al. 2017; Prates et al. 2020), were collected at the northern region of the Central Cordillera in the municipality of Anori. Therefore, neither samples from the species’ type locality nor its surroundings in the Western Cordillera have been included in molecular phylogenetic analyses so far, and consequently, the phylogenetic affinities between western and central populations remain entirely unknown. Velasco and Hoyos (2010) noted that the morphological traits of individuals assigned to Anolis danieli from the municipality of Anori, Antioquia do not match the diagnostic characters of the species’ type series. We confirm this observation through our own examination of specimens from Anori, including the voucher specimens used in the phylogeny of Castañeda and de Queiroz (2011), as well as a reexamination of the type series of A. danieli. These observations indicate that the sample analyzed by Castañeda and de Queiroz (2011) were misidentified as A. danieli. Therefore, the phylogenetic position of this species within the latifrons series remains poorly understood. Additionally, the phylogenetic position of other species, including A. limon Velasco & Hurtado-Gómez, 2014 and A. mirus Williams, 1963, have been assessed solely based on morphological data, leaving their placement in molecular phylogenies uncertain. In this study, we combined both published and newly generated molecular data to reconstruct the phylogeny of the latifrons series and test the phylogenetic positions of two species for which previous phylogenies incorporated only morphological information. We also gathered morphological and geographical data to formally describe the taxon previously assumed to be A. danieli. We expect this study to clarify the phylogenetic relationships, morphological variation, and geographic distribution of species in the latifrons series, thereby contributing to a broader understanding of the biodiversity patterns within the highly diverse Dactyloa clade. Methods Institutional acronyms We examined the available material of the latifrons series housed in the following Colombian herpetological collections: Museo de Herpetología Universidad de Antioquia (MHUA), Colecciones Biológicas Universidad CES (CBUCES), Museo de Ciencias Naturales de La Salle (CSJ), and Colección Biológica Universidad EAFIT (EAFIT). Vertebrate Zoology 75, 2025, 441–457 443 Taxon sampling We examined specimens of Anolis apollinaris Boulenger, 1919, A. danieli (including the holotype), A. fraseri, A. frenatus, A. latifrons, A. limon (including the holotype), A. maculigula Williams, 1984, A. princeps Boulenger, 1902, A. purpurescens Cope, 1899, an undescribed species previously allocated to the latifrons series (i.e., A. sp.2 from Castañeda and de Queiroz 2011) and specimens identified to the genus level (i.e., Anolis sp.). In addition, we included the only known (to our knowledge) specimens of A. mirus beyond its type locality in the western region of Colombia. In total we gathered morphometric and meristic data for 200 specimens. From these specimens we generated genetic data for the following species (number of sequenced specimens in parentheses): A. maculigula (2), A. latifrons (4), A. frenatus (3), the lineage labeled as A. danieli by Castañeda and de Queiroz (2011) (8), the true A. danieli (12), A. purpurescens (1) and, for the first time within the Dactyloa clade, we generated sequences for both A. limon (2) and A. mirus (1). DNA sequence data and phylogenetic analyses We amplified and sequenced: a partial region of the mitochondrial ribosomal unit 16S using the primers 16SCL-16SDH (Santos et al. 2009) and 16Sar-16Sbr (Santos et al. 2003), a partial region of the cytochrome oxidase subunit 1 COI using the primers repCOI-F-repCOI-R (Palumbi 1996), and a partial region of the NAD dehydrogenase subunit 2 ND2 using the primers metf6-L5556 (Castañeda and de Queiroz 2011). We combined these newly generated sequences with available sequences for these three regions and the nuclear regions ECEL1 and rag1 from GenBank (www.ncbi. nlm.nih.gov/Genbank) and the Steven Poe’s webpage (https://www.stevenpoe.net). In total for the latifrons series, we obtained 199 sequences (65 generated by us and 134 gathered from previous studies), representing 22 out of the 25 named species of the group (88% of the its known richness). For the remaining species, namely Anolis propinquus Williams, 1984, A. squamulatus Peters, 1863 and A. savagei Poe & Ryan, 2017, no genetic data are available; therefore, they were not included in our assembled dataset. We aligned each region using MAFFT under default parameters (Katoh and Standley 2013). We assembled a single matrix and used ModelFinder (Kalyaanamoorthy et al. 2017) to determine the best partition scheme and the evolutionary model for each partition. We inferred a phylogenetic tree using Maximum Likelihood as implemented in IQTREE 2 (Minh et al. 2020). Nodal support was estimated with 5000 pseudoreplicates using the ultrafast bootstrap approach (Minh et al. 2013). We then estimated raw genetic distances using the 16S fragment (the region with more sequences available) using the dist.dna function in the ape R package (Paradis et al. 2004). Morphological data We assembled a dataset with 24 morphometric variables: 16 linear and eight meristic. Linear measurements and meristic traits were selected following Köhler (2014) and Williams et al. (1995) respectively. In total, we obtained morphometric data from 200 specimens corresponding to ten formally described species and two undescribed taxa according to Castañeda and de Queiroz (2011), namely, Anolis sp.1 and A. sp.2. Our dataset also includes 42 samples of Anolis danieli from across its geographic distribution (33 samples from the Western Cordillera and 9 from the Central Cordillera). Linear variables are: HL (head length), HW (head width), HH (Head height), SL (snout length), LDE (length of the meatus), LDP (length of the interparietal scale), HuL (humerus length), UlL (ulnae length), FL (femur length), FoL (foot length), ShL (shank length), H (hand length), Fourth_L (fourth toe length), SPW (subdigital pad width), TL (tail length) and SVL (snout-vent length). We measured H and FL since these variables were used by Ayala-Varela (2021) to compare A. nemontae with its related species A. fraseri. Meristic variables are: Canths_2 (number of scales between second canthals), PR (number of postrostral scales), IO (number of scales between supraorbital semicircles), IP/ IO (number of scales between interparietal and semicircles), SPLeye (number of supralabial scales to the level below center of the eye), PM (number of postmental scales), ToeLam2_3 (Counting the plates under phalanges II–III) and ToeLam2_4 (Counting the plates under phalanges II–IV). We counted this latter character following both Williams et al. (1995) and Köhler (2014) to facilitate species comparisons, as the nomenclature of this trait varies across studies. Linear measurements were taken under the stereo microscope with a digital caliper to the nearest 0.1 mm. Measurements of museum vouchers examined in this study and associated geographic data are available in the File S1. Sex was determined by the presence of hemipenis and, in some species, by the presence of enlarged postcloacal scales in males. Dewlap size was also used as a sexually dimorphic character: in males, the dewlap extends posteriorly beyond the level of the forelimb insertion, whereas in females it does not. The coloration description was based on field observations and photos of live specimens. Comparisons with congeneric species of the latifrons series were conducted by a detailed examination of original descriptions and specimens housed in the herpetological collections visited (see the Taxon sampling section). Morphometric analyses To evaluate the morphologic distinctiveness among species of the latifrons series, we conducted a discriminant analysis coupled to a principal component analysis (DAPC; Jombart et al. 2010) on our assembled measurements and meristic dataset. DAPC is a supervised method that utilizes a linear combination of intercorrelated Marin CM et al.: Anolis Systematics 444 descriptors to maximize between-group variance while minimizing within-group variance, allowing individuals from predefined species to be assigned to distinct clusters. Implementation of DAPC is preferable over other multivariate approaches such as PCA alone, which focuses on trait covariation but ignores the cluster identity of individuals, potentially overlooking actual differences among groups (Jombart et al. 2010). Nineteen individuals lacked linear measurements for TL due to incomplete tails. Thus, to include these specimens in our morphometric analysis, we employed a missing data imputation approach. Specifically, we used the mean value for this continuous trait calculated across all individuals of a given species and considering only adult specimens. We conducted the DAPC analysis using species identity as the output variable and morphometric values as the input variables. To estimate the minimum number of principal components accounting for the highest morphological covariation and predicted cluster membership, we used crossvalidation with 90% of the data for training, 10% for testing, and 300 replicates at each level of PCA retention. Cross-validation analysis aims at selecting the number of principal components (PCs) that maximize the mean success assignment while minimizing the root square error of assignments (e.g., <1% of successful rate assignment) or the lower number of PCs. The optimal number of PCs was selected through cross-validation, and we retained the lowest number of discriminant functions that allowed classification success rates above 90%. To determine which type of dataset, based on either linear measurements or meristic data, most effectively distinguishes among the examined species, we performed DAPC analyses on three independent datasets: 1 – only linear measurements (LM), 2 – only meristic data (MD), and 3 – a combined dataset (CD) containing both linear measurements and meristic data. Our morphometric analyses included only adult specimens from species with measurements available for more than two individuals. Cross-validation and DAPC analyses were conducted using the adegenet R package (Jombart 2008; R Core Team 2025). Results Phylogenetic inference The assembled matrix included 235 terminals and 5533 segregating sites. The best partition scheme and substitution models were: GTR+F+I+R4 (16S), SYM+I+G4 (COI_1), F81+F+I+R2 (COI_2), TIM2+F+R7 (COI_3+- ND2_3), TR+F+I+R5 (ND2_1), TVM+F+I+R5 (ND2_2), and TIM3+F+R2 (ECEL1+RAG1). The obtained phylogenetic tree was well supported with most of the nodes with UFB>95% (Fig. 1). Although our phylogenetic tree represents the most comprehensive analysis of the latifrons series to date, many of the relationships we recovered are consistent with the hypothesis proposed by Poe et al. (2017), which integrated both morphological and genetic data (Fig. 2). Minor topological differences with respect to previous studies involve species for which we generated novel sequences, and these are highlighted as follows: Anolis limon was recovered as the sister species to the clade comprising A. purpurescens, A. ibanezi Poe, Latella, Ryan & Schaad, 2009, and A. maia (Batista et al., 2015), rather than as the sister to the clade comprising A. purpurescens and A. ibanezi, as reported by Poe et al. (2017). Our results also differ from those of Batista et al. (2015), who recovered A. maia as sister to the clade comprising A. ibanezi and A. purpurescens, whereas our analysis supports A. maia as the sister species to A. ibanezi. Likewise, our tree recovers A. mirus as the sister taxon to the clade comprising A. parilis Williams, 1975, A. fraseri, and A. nemontae Ayala-Varela et al., 2021, in contrast to Poe et al. (2017), who recovered A. mirus as the sister species of A. kunayalae Hulebak, Poe, Ibáñez & Williams, 2007 (Figs 1, 2). A full tree including 235 anole species and the corresponding matrix alignment and partitions scheme are available in Files S2–S4 respectively. ENA/GenBank accession numbers for species of the latifrons series used in this study including novel generated sequences are available in the File S1. Our most significant finding, which contrasts with previous hypotheses, concerns the phylogenetic position of Anolis danieli. Castañeda and de Queiroz (2011) included genetic samples identified as A. danieli. However, upon examining those specimens, reviewing the holotype and paratype, and incorporating additional specimens and genetic data, we determined that the samples used in previous studies represent an undescribed taxon. We also included genetic data from specimens we consider to be the true A. danieli, collected from multiple localities, including sites near the type locality. Our analyses recovered A. danieli as sister, with 100% UFB support, to a clade comprising A. purpurescens, A. maia, A. ibanezi, and A. limon (Figs 1, 2). The new taxon, previously misidentified as A. danieli, was found to be sister to another undescribed species from the Santander department (Fig. 1). The genetic sample labeled as A. sp.1 by Castañeda and de Queiroz (2011) was found to be nested within the clade representing the true A. danieli. In addition, our molecular phylogeny included A. brooksi and the recently described species Anolis kathydayae Poe & Ryan, 2017 but found them to show no evidence of genetic differentiation (Figs 1–3). Overall, the interspecific genetic distances in the latifrons series ranged from 2% to 9% (Fig. 3). However, a particular case involves the genetic distances among A. insignis Cope, 1871, A. kathydayae, and A. brooksi. In the inferred phylogeny, A. insignis is clearly separated from the clade comprising (A. kathydayae + A. brooksi). Although there is an imbalance in genetic marker representation for these terminals (File S3), analysis of the COI region shows an uncorrected genetic distance of 10% between A. insignis (POE2015) and A. brooksi (File S5). In contrast, analysis of the 16S region reveals an uncorrected genetic distance of 0% between the Vertebrate Zoology 75, 2025, 441–457 445 0.04 Poe2015 MHUAR13667 SMF91459 MHCH2784 MRC135 SMF89482 QCAZ6868 SMF91499 SMF91455 MHCH 2788 MHUAR13262 SMF91501 MHUAR11558 MHUAR12976 SMF96576 JMS192 POE2017 SMF91504 SMF91477 Poe2015 MHUAR13268 MHUAR11559 SMF96579 MHCH2787 Poe2015 MVUP2021 SMF90098 MHUAR12987 SMF89737 MHUAR12904 MHCH2637 JMS214 Poe2015 MHCH2786 KEN20042 SMF91460 SMF91453 POE4407 SMF96577 MHUAR13677 SMF89453 SMF89496 MHCH2120 QCAZ6892 MHUAR13525 SMF91500 6 9 87 ginaelisae microtus insignis kathydayae brooksi agassizi MHUAR12906 apollinaris maculigula casildae princeps latifrons frenatus CH6023 CH5833 CH5813 CH5217 72 to Fig. 1B 1A 0.04SMF97270 MHUAR12969 QCAZ14317 MHUAR11567 MHUAR13660 QCAZ15058 MHUAR11565 QCAZ8025 MHUAR11455 SMF97271 MHUAR13754 CH7699 SMF97272 MHCH2782 QCAZ15386 MHUAR13542 QCAZ6862 MHUAR13756 QCAZ6867 SMF91485 POE1966 MHUAR13755 MHUAR13061 SMF89459 CH6031 MHUAR13126 MHUAR11570 QCAZ14431 MHUAR13753 SMF97269 MHUAR11564 SMF91476 MRC123 CSJH5025 MHUAR11595 Poe2017 QCAZ3441 MHUAR12625 CH8461 MRC134 MHUAR12749 MHUAR13535 MHUAR13253 MHUAR13659 MHCH2635 MHUAR13757 MHUAR13421 MHUAR11562 QCAZ14393 QCAZ14596 MHUAR13540 MHUAR12834 MHUAR11594 MHCH2783 MHUAR12510 MHUAR11571 81 87 7 1 91 94 ibanezi maia nemontae danieli limon purpurescens mirus kunayalae cyanophthalmus sp. nov. sp.2 fraseri parilis 1B Figure 1. A, B Phylogenetic relationships among species of the latifrons series as inferred from molecular data. The geographic distribution of each lineage is depicted, with the colors of the tree tips corresponding to the dots on the map. Geographic occurrences for species not included in our main morphological dataset were taken from Velasco et al. (2020). Dots colored with white marks in the center represent sequenced voucher included in the phylogeny. Locality record for A. savagei is indicated on the map but this species was not included in the phylogeny since no molecular data are available. Star symbol represents the type locality of Anolis cyanophthalmus sp. nov. Marin CM et al.: Anolis Systematics 446 same species (Fig. 3). While it is well established that COI evolves faster than 16S, such an extreme discrepancy between the two markers is unexpected. Notably, A. insignis (tip labeled POE2015 in Fig. 1) is not associated with a voucher specimen, and its sequence is not available in GenBank; we retrieved it instead from the supplementary material of Poe et al. (2017). Therefore, the interspecific relationships shown in our phylogeny casildae limon kunayalae insignis cyanophthalmus sp. nov. apollinaris maia sp. 2 ginaelisae maculigula brooksi nemontae mirus princeps danieli latifrons purpurescens kathydayae parilis ibanezi frenatus fraseri microtus agassizi squamulatus propinquus Figure 2. Comparison of our molecular phylogeny with the comprehensive phylogeny of the latifrons series by Poe et al. (2017). Blue tips indicate lineages for which we added genetic sequences for the first time. Red tips represent those species for which no molecular data are available. Green numbers indicate nodes present in Poe et al. (2017) but with lower support values. 0.000 0.025 0.050 0.075 Genetic distance 0.00 0.08 0.08 0.07 0.07 0.03 0.07 0.00 0.00 0.07 0.08 0.07 0.09 0.08 0.03 0.08 0.09 0.09 0.01 0.07 0.05 0.06 0.09 0.06 0.05 0.07 0.06 0.07 0.06 0.04 0.06 0.09 0.06 0.07 0.07 0.00 0.05 0.08 0.08 0.05 0.04 0.06 0.06 0.08 0.05 0.08 0.06 0.09 0.07 0.06 0.05 0.01 0.04 0.07 0.02 0.04 0.06 0.04 0.05 0.02 0.07 0.02 0.07 0.04 0.02 0.05 0.00 0.07 0.05 0.04 0.06 0.05 0.02 0.05 0.07 0.05 0.07 0.05 0.06 0.08 0.00 0.06 0.01 0.02 0.07 0.08 0.06 0.09 0.07 0.03 0.08 0.09 0.09 0.01 0.05 0.06 0.04 0.05 0.02 0.07 0.02 0.07 0.04 0.02 0.06 0.00 0.00 0.05 0.05 0.04 0.06 0.04 0.03 0.05 0.06 0.04 0.00 0.06 0.07 0.06 0.08 0.06 0.03 0.07 0.07 0.07 0.01 0.06 0.04 0.06 0.04 0.07 0.02 0.04 0.06 0.01 0.05 0.07 0.06 0.08 0.06 0.06 0.08 0.00 0.07 0.02 0.07 0.04 0.03 0.06 0.00 0.08 0.09 0.07 0.08 0.07 0.00 0.08 0.04 0.02 0.07 0.00 0.08 0.08 0.08 0.00 0.04 0.07 0.00 0.07 0.00 sp. (Santander) purpurescens mirus microtus maia maculigula limon latifrons kunayalae kathydayae insignis ibanezi ginaelisae frenatus danieli cyanophthalmus casildae brooksi sp. (Santander) purpurescens mirus microtus maia maculigula limon latifrons kunayalae kathydayae insignis ibanezi ginaelisae frenatus danieli cyanophthalmus casildae brooksi Figure 3. Heat map of estimates of net uncorrected genetic distances between species of the latifrons series as estimated using sequences of the 16S fragment. Note the lack of genetic divergence (light blue) between samples of A. brooksi and A. kathydayae. Vertebrate Zoology 75, 2025, 441–457 447 for this small subclade were mainly resolved by the COI region. Morphological analysis After filtering for adults only, the final dataset used for the morphometric analyses included 109 individuals. The DAPC analysis successfully assigned between 91% (MD and CD datasets) and 99% (LM dataset) of these individuals to the appropriate species, including samples of the true Anolis danieli and the new taxon (Fig. 4) by summarizing the original morphometric variables into 13 PCs in the LM dataset, 6 PCs in the MD dataset, and 16 PCs in the CD dataset. These principal components were then combined into 6 discriminant functions (DFs) in both the LM and MD datasets, and eight DFs in the CD dataset. Most misclassification (nine) occurred in the analysis of the MD dataset, where several species showed substantial overlap. In contrast, using both the LM and CD datasets resulted in the misclassification of only one individual. Results from the LM dataset revealed morphometric differences among species primarily related to the variables LDE (0.58), HH (0.13), and LDP (0.10). Most differences in both the MD and CD datasets were associated with the variable IP/IO (0.43 and 0.33, respectively). In the LM dataset, the second discriminant function showed the strongest correlations with the variable HL (0.31), while in the MD and CD datasets, it was strongly correlated with the variables ToeLam2_3 and PR (0.39 and 0.15 respectively), as well as with PM (0.34 and 0.12 respectively). All this morphometric variation was independent of body size (SVL), which contributed less than 0.6% to each discriminant function. Since both the LM and CD datasets were quantitatively similar, we present our results based on the LM dataset (Fig. 4). Anolis frenatus Anolis latifrons Anolis cyanophthalmus sp. nov. Anolis danieli Anolis limon Anolis maculigula Anolis purpurescens Figure 4. Morphometric differences among species of the latifrons series inferred from a DAPC analysis using the LM dataset. The top panel shows the posterior probability of individual assignment to each species (bars), relative to their actual species identity (filled circles). The bottom panel displays individuals in the morphometric space as represented by the first two discriminant functions. Inertia ellipses are included to facilitate visualization of group clustering. Marin CM et al.: Anolis Systematics 448 Discussion In this study, we integrated historical and newly generated genetic data, including the first available sequences for two species in the latifrons series to reconstruct the most comprehensive molecular phylogeny of the clade to date. Based on our phylogenetic results, we tested the placement of these species and evaluated the species boundaries of two recently described Central American taxa. Through a thorough morphological examination of museum specimens, we determined that the taxon labeled as Anolis danieli in previous phylogenies was misidentified and actually corresponds to an undescribed species. In addition, we expanded the geographic and genetic sampling for the true A. danieli within the phylogeny of the latifrons series. Below, we discuss our main findings in detail. Phylogenetic position of Anolis danieli, A. limon and A. mirus Phylogenetic studies that have addressed the evolutionary relationships within the latifrons series have included either morphological or molecular samples of the nominal species Anolis danieli (Velasco and Hoyos 2010; Castañeda and de Queiroz 2011, 2013). In this study, we conducted a phylogenetic analysis of the latifrons series, including genetic samples of A. danieli from across its known geographic range in the Western Cordillera, and analyzed these samples along with those from the municipality of Anori and other localities in the northern Antioquia, including samples used in all previous molecular phylogenies of the latifrons series. Strikingly, our analyses showed that samples assigned to A. danieli by Castañeda and de Queiroz (2011) do not actually correspond to this species. Instead, they represent an undescribed taxon, which had been consistently misidentified in prior phylogenetic studies (Castañeda and de Queiroz 2011, 2013; Prates et al. 2017; Prates et al. 2020). The phylogenetic position of the new taxon, as sister to an undescribed species from the Eastern Cordillera in the department of Santander (A. sp.2 sensu Castañeda and de Queiroz 2011), is consistent with prior hypotheses (Castañeda and de Queiroz 2011, 2013; Prates et al. 2017, 2020). However, unlike earlier studies, our analyses retrieved A. danieli (i.e., A. sp.1, sensu Castañeda and de Queiroz 2011) as sister to the remaining green anoles from the midlands of the Central Cordillera and the lowlands of the Pacific region and Panama (Fig. 1). Our results also help clarify the phylogenetic positions of two species that had previously been placed in the phylogeny of the group based solely on morphology due to the absence of molecular data. The first species, Anolis limon, was initially proposed by Velasco and Hurtado-Gómez (2014) as the sister species to A. ibanezi. Later, Poe et al. (2017), in their systematic assessment of the entire genus Anolis, recovered A. limon as the sister species to the clade comprising A. ibanezi and A. chocorum (A. purpurescens). Contrary to these previous hypotheses, our phylogeny recovered A. limon as sister to the clade comprising A. purpurescens, A. ibanezi, and A. maia. The second species, Anolis mirus, was suggested to be part of the aequatorialis series by Williams (1975), Ayala-Varela and Velasco (2010), and Castañeda and de Queiroz (2013) although none of these studies conducted a formal test of its phylogenetic position. Poe et al. (2017) later placed A. mirus within the latifrons series as sister to A. kunayalae. However, contrary to these earlier hypotheses, our analysis recovered A. mirus as the sister species to the clade comprising A. parilis, A. nemontae, and A. fraseri. Evaluating the utility of morphological data for the taxonomy of the latifrons series The LM dataset performed the best in distinguishing among species of the latifrons series. Overall, the cluster formed by A. danieli, A. limon, and A. purpurescens showed significant overlap in the morphometric space. Although we were unable to include data for A. ibanezi, its morphological similarity to the remaining species of the clade in which it is nested suggests that the observed overlap pattern would likely persist with its inclusion in morphometric analysis. Anolis frenatus and A. latifrons also showed substantial morphometric overlap, likely reflecting their similarity in body size and limb proportions. Likewise, although we did not include samples from A. princeps, we suspect this species would also exhibit a high degree of morphometric overlap with the cluster comprising A. frenatus and A. latifrons, likely reflecting their conserved external morphology and phylogenetic proximity. Indeed, this similarity has led previous authors to question the validity of these taxa as distinct species (Savage and Talbot 1978; Williams 1988). In contrast, the MD dataset did not allow for a clear distinction among the species of the latifrons series, with the notable exception of Anolis maculigula, which strongly segregates in the morphometric space across all datasets. This is likely due to A. maculigula possessing the highest number of scales between the second canthals among the species examined (Table 1). Based on these findings, we recommend future morphometric studies of the latifrons series to prioritize the analysis of linear measurements, as meristic traits alone seem to offer limited resolution for distinguishing among species in this group. Recent morphometric assessments of the Andean (Grisales-Martínez et al. 2017) and Central American fuscoauratoids (Köhler and Sunyer 2008; Ponce and Köhler 2008) have reached similar conclusions, suggesting that meristic data provide limited performance in morphometric analyses of mainland anoles. Nonetheless, these results should be interpreted with caution, as our study lacks morphometric information for almost 60% of the known species in the group. Vertebrate Zoology 75, 2025, 441–457 449 Biogeography Our reconstructed phylogeny places the highland species Anolis danieli as sister to the remaining green anoles distributed in the mid-elevation areas of the Magdalena Valley (A. limon), the Pacific lowlands of Colombia (A. purpurescens), and the lowlands of Panama (A. ibanezi and A. maia). This branching pattern suggests that diversification within the clade occurred along a highland-lowland axis, potentially involving an ancestor inhabiting the highlands of the Western Cordillera. Subsequent downslope range expansion and dispersal toward the western slope of the Central Cordillera, and ultimately into the Pacific region and Panama, may have driven a rapid radiation of the midto lowland subclade. A highland origin followed by downslope diversification has also been hypothesized for other anoles and frogs from montane regions of South America (Firkowski et al. 2016; Prates et al. 2020). Moreover, this scenario may help explain the close phylogenetic relationships, low genetic distances, and morphometric overlap observed among species in this subclade. The species previously misidentified as Anolis danieli and its sister species (A. sp.2 sensu Castañeda and de Queiroz 2011) are allopatrically distributed, with the former occurring on the eastern slope of the Central Cordillera between 1500 and 1900 m a.s.l. and the latter on the western slope of the Eastern Cordillera at 1360 m a.s.l. approximately (Castañeda and de Queiroz 2011). During the late Miocene, when the divergence of both species is estimated to have occurred (Prates et al. 2017, 2020), the Eastern Cordillera underwent a dynamic uplift phase, reaching elevations between 1600 and 2100 m a.s.l. (Boschman 2021). Such intensive orogenetic activity may have influenced the divergence of these species by splitting the range of a putative ancestral population, a mechanism also implicated in the diversification of the highland Andean anoles of the heterodermus series (Moreno-Arias et al. 2023). Table 1. Comparison of meristic characters used in the taxonomy of the latifrons series sensu Poe et al. (2017) including the new species we described herein (see main text for details). Sources: 1– This study, 2 – Arosemena, Ibáñez & de Sousa, 1991, 3 – Ayala et al. (2021), 4 – Batista et al. (2015), 5 – Hulebak et al. (2007), 6 – Lotzkat et al. (2013), 7 – Poe and Ryan (2017), 8 – Poe et al. (2009), 9 – Rivero et al. (2009), 10 – Stejneger (1900), 11 – Velasco and Hurtado-Gómez (2014), 12 – Williams (1963), 13 – Williams (1975), 14 – Williams (1984), 15 – Williams (1984), 16 – Williams (1988) and 17 – Zambrano et al. (2024). Values in bold correspond to new data on the range of variation collected in this study; the remaining data were gathered from literature sources. The complete variable names are provided in the Morphological data section. Species Canths_2 PR IO IP/IO SPLeye PM ToeLam2_3 ToeLam2_4 Source A. cyanophthalmus sp. nov. 9–14 4–9 1–3 3–6 7–10 6–9 22–30 35–47 1 A. agassizi – – – – – – 36 – 10 A. apollinaris 8–16 8 2–4 2–5 7–8 5 23–29 41 1, 9, 16 A. brooksi 10–11 6–7 3–4 2–4 7–9 5–7 25–28 – 7 A. casildae 11–17 6–11 2–5 2–5 6–8 6–9 23 31–42 2, 6 A. danieli 8–14 4–10 1–5 2–7 6–9 5–8 19–28 29–43 1, 16 A. fraseri 6–10 5–9 2–5 2–5 6–9 4–8 18–24 28 3, 16 A. frenatus 9–15 5–9 2–5 3–6 8–12 6–11 22–30 35–43 1, 2, 6, 16 A. ginaelisae 5–9 4–8 1–3 3 6–8 4–7 – 29–36 6 A. ibanezi 8–12 6–8 2–3 2–3 7–8 6–8 17–19 27–30 4, 6, 8 A. insignis 9–11 7–10 2–3 2–3 8–9 6–9 25–27 37–41 7 A. kathydayae 9–11 5–6 3–4 3–4 7–8 4–5 23–27 – 7 A. kunayalae 12–17 6–9 4–5 3–6 7–10 6–8 11–15 18–24 4, 5, 6 A. latifrons 11–15 6–9 3–5 4–8 8–11 6–10 21–27 36–43 1, 4, 16 A. limon 8–12 5–9 1–3 1–5 8–10 6–8 19–25 28–36 1, 11,17 A. maculigula 12–19 7–11 1–5 0–3 6–10 4–12 16–23 29–35 1, 15 A. maia 10–16 6–8 2–3 3–5 9–11 6–7 – 28– 32 4 A. microtus 6–9 5–7 1–2 1 7–9 6–8 20–22 32–34 2, 6, 16 A. mirus 12–14 6–7 4–5 3–4 7–10 6–8 15–18 25–26 1, 12, 13 A. nemontae 7–11 6–9 3–4 2–3 7–10 5–7 21–23 – 3 A. parilis 17 6 4 6 9 8 15 – 13 A. princeps 12–17 5–9 2–6 3–7 9–12 6–11 22–25 39–40 1, 16 A. purpurescens 7–15 5–10 1–5 2–6 7–10 1–7 18–21 25–33 1, 4 A. propinquus 12 6 3 – 7–8 6 25 – 14, 16 A. savagei 8–9 6–7 2 1–2 7 7–8 25–29 – 7 A. squamulatus 11–15 – 3–5 4–8 6–11 – 22–27 – 16 Marin CM et al.: Anolis Systematics 456 Palumbi SR (1996) Nucleic acids II: The polymerase chain reaction. In: Hillis DM, Moritz C, Mable BK (Eds) Molecular Systematics, Second Edition. Sinauer Associates, Sunderland, MA, 205–221. Paradis E, Claude J, Strimmer K (2004) APE: Analyses of phylogenetics and evolution in R language. Bioinformatics 20: 289–290. https://doi.org/10.1093/bioinformatics/btg412 Poe S (2004) Phylogeny of anoles. Herpetological Monographs 18: 37– 89. https://doi.org/10.1655/0733-1347(2004)018[0037:POA]2.0.CO;2 Poe S, Latella I, Ayala-Varela F, Yañez-Miranda C, Torres-Carvajal O (2015) A new species of phenacosaur Anolis (Squamata; Iguanidae) from Peru and a comprehensive phylogeny of Dactyloa-clade anolis based on new DNA sequences and morphology. Ichthyology & Herpetology 103: 639. https://doi.org/10.1643/CH-14-127 Poe S, Nieto-Montes de Oca A, Torres-Carvajal O, de Queiroz K, Velasco JA, Truett B, Gray LN, Ryan MJ, Köhler G, Ayala-Varela F, Latella I (2017) A phylogenetic, biogeographic, and taxonomic study of all extant species of Anolis (Squamata; Iguanidae). Systematic Biology 66: 663–697. https://doi.org/10.1093/sysbio/syx029 Poe S, Ryan MJ (2017) Description of two new species similar to Anolis insignis (Squamata: Iguanidae) and resurrection of Anolis (Diaphoranolis) brooksi. Amphibian & Reptile Conservation 11: 1–16. Ponce M, Köhler G (2008) Morphological variation in anoles related to Anolis kemptoni in Panama. Salamandra 44: 65–83. Prates I, Melo-Sampaio PR, Drummond L de O, Teixeira M, Rodrigues MT, Carnaval AC (2017) Biogeographic links between southern Atlantic Forest and western South America: Rediscovery, re-description, and phylogenetic relationships of two rare montane anole lizards from Brazil. Molecular Phylogenetics and Evolution 113: 49– 58. https://doi.org/10.1016/j.ympev.2017.05.009 Prates I, Melo-Sampaio PR, Queiroz K De, Carnaval AC, Rodrigues MT, Drummond LDO (2020) Discovery of a new species of Anolis lizards from Brazil and its implications for the historical biogeography of montane Atlantic Forest endemics. Amphibia Reptilia 41: 87–103. https://doi.org/10.1163/15685381-20191179 R Core Team (2025) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna. https:// www.R-project.org Rivero EI, Velozo P, Rojas-Runjaic F (2009) Primer registro del lagarto Anolis apollinaris, para Venezuela. Boletín del Centro de Investigaciones Biológicas 43: 299–304. Santos JC, Coloma LA, Cannatella DC (2003) Multiple, recurring origins of aposematism and diet specialization in poison frogs. Proceedings of the National Academy of Sciences of the USA 100: 12792–12797. https://doi.org/10.1073/pnas.2133521100 Santos JC, Coloma LA, Summers K, Caldwell JP, Ree R, Cannatella DC (2009) Amazonian amphibian diversity is primarily derived from Late Miocene Andean lineages. PLoS Biology 7: e1000056. https://doi.org/10.1371/journal.pbio.1000056 Savage JM, Guyer C (1989) Infrageneric classification and species composition of the anole genera, Anolis, Ctenonotus, Dactyloa, Norops and Semiurus (Sauria: Iguanidae. Amphibia-Reptilia 10: 105–116. Savage JM, Talbot JJ (1978) The giant anoline lizards of Costa Rica and western Panama. Copeia 1978: 480–492. https://doi.org/ 10.2307/1443615 Torres-Carvajal O, Ayala-Varela FP, Lobos SE, Poe S, Narváez AE (2018) Two new Andean species of Anolis lizard (Iguanidae: Dactyloinae) from southern Ecuador. Journal of Natural History 52: 1067–1089. https://doi.org/10.1080/00222933.2017.1391343 Uetz P, Freed P, Aguilar R, Reyes F, Hošek J (Eds) (2023) The Reptile Database. http://www.reptile-database.org [accessed 5 Jun 2025]. Velasco JA, Hoyos JM (2010) Phylogenetic analysis of the latifrons series of Anolis (Sauria: Iguania) with morphological data. Anolis Newsletter VI: 203–212. Velasco JA, Hurtado-Gómez JP (2014) A new green anole lizard of the Dactyloa clade (Squamata: Dactyloidae) from the Magdalena River valley of Colombia. Zootaxa 3785: 201–216. https://doi. org/10.11646/zootaxa.3785.2.4 Velasco JA, Villalobos F, Diniz-Filho JAF, Poe S, Flores-Villela O (2020) Macroecology and macroevolution of body size in Anolis lizards. Ecography 43: 812–822. https://doi.org/10.1111/ecog.04583 Williams EE (1975) South American Anolis: Anolis parilis, new species, near A. mirus. Breviora 434: 1–8. Williams EE (1976) South American anoles: The species groups. Papéis Avulsos de Zoologia 29: 259–268. Williams EE (1988) New or problematic Anolis from Colombia. V. Anolis danieli, a new species of the latifrons species group and a reassessment of Anolis apollinaris Boulenger, 1919. Breviora 489: 1–25. Williams EE, Rand H, Rand S, O’Hara R (1995) A computer approach to the comparison and identification of species in difficult taxonomic groups. Breviora 502: 1–47. Yánez-Muñoz MH, Reyes-Puig C, Reyes-Puig JP, Velasco JA, Ayala-Varela F, TorresCarvajal O (2018) A new cryptic species of Anolis lizard from northwestern South America (Iguanidae, Dacty loinae). ZooKeys 2018: 135–163. https://doi.org/10.3897/zookeys.794.26936 Zambrano DF, Gutierrez W, Zuñiga-Baos J, Cano K, Parra RM, Bernal MH (2024) New distribution records, morphology and natural history notes of the endemic Colombian lizard Anolis limon. Herpetological Bulletin 169: 7–11. https://doi.org/10.33256/hb169.711 Vertebrate Zoology 75, 2025, 441–457 457 Supplementary Material 1 Files S1–S5 Authors: Marín CM, Bocanumenth D, Daza JM (2025) Data type: .zip Explanation notes: File S1. Specimen data including museum, geographic, ENA/GenBank accesion numbers and morphometric data [.xlsx file]. — File S2. Complete phylogenetic tree in newick format including 235 terminals (outgroups) [.contree file]. — File S3. As sembled genetic dataset incuding five genomic regions [.phy file]. — File S4. Nexus file describing the genomic partitions used in the phylogenetic analysis [.nex file]. — File S5. Heat map of net uncorrected genetic distances between species of the latifrons series as estimated using sequences of the COI fragment [.pdf file]. Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons.org/licenses/ odbl/1.0). The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited. Link: https://doi.org/10.3897/vz.74.e162071.suppl1