https://orcid.org/0000-0002-0242-6385
Figures
Abstract
The eyes of insects display an incredible diversity of adaptations to enhance vision across the gamut of light levels that they experience. One commonly studied contrast is the difference in eye structure between nocturnal and diurnal species, with nocturnal species typically having features that enhance eye sensitivity such as larger eyes, larger eye facets, and larger ocelli. In this study, we compared eye structure between workers of closely related nocturnal and diurnal above ground foraging ant species (Hymenoptera: Formicidae) in four genera (Myrmecocystus, Aphaenogaster, Temnothorax, Veromessor). In all four genera, nocturnal species tend to have little cuticular pigment (pale), while diurnal species are heavily pigmented (dark), hence we could use cuticle coloration as a surrogate for activity pattern. Across three genera (Myrmecocystus, Aphaenogaster, Temnothorax), pale species, as expected for nocturnally active animals, had larger eyes, larger facet diameters, and larger visual spans compared to their dark, more day active congeners. This same pattern occurred for one pale species of Veromessor, but not the other. There were no consistent differences between nocturnal and diurnal species in interommatidial angles and eye parameters both within and among genera. Hence, the evolution of eye features that enhance sensitivity in low light levels do not appear to have consistent correlated effects on features related to visual acuity. A survey across several additional ant genera found numerous other pale species with enlarged eyes, suggesting these traits evolved multiple times within and across genera. We also compared the size of the anterior ocellus in workers of pale versus dark species of Myrmecocystus. In species with larger workers, the anterior ocellus was smaller in pale than in dark species, but this difference mostly disappeared for species with smaller workers. Presence of the anterior ocellus also was size-dependent in the two largest pale species.
Citation: Johnson RA, Rutowski RL (2022) Color, activity period, and eye structure in four lineages of ants: Pale, nocturnal species have evolved larger eyes and larger facets than their dark, diurnal congeners. PLoS ONE 17(9): e0257779. https://doi.org/10.1371/journal.pone.0257779
Editor: Nicolas Chaline, Universidade de São paulo, BRAZIL
Received: September 8, 2021; Accepted: August 24, 2022; Published: September 22, 2022
Copyright: © 2022 Johnson, Rutowski. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All data are deposited in the Dryad Digital Repository at doi:10.5061/dryad.nzs7h44v8.
Funding: The authors received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Ectotherms display extensive variation in color that arises at least in part from variation in the amount of pigment deposited in the cuticle/integument, with melanin being the most common pigment [1, 2]. Diverse selective factors favor the evolution of dark body coloration including biotic factors such as predation, sexual selection, and disease resistance [3–8], and abiotic factors such as temperature, ultraviolet–B radiation, and desiccation [9–13]. Despite the various potential benefits of melanin deposition, numerous clades contain species with little or no pigment in their integument. These pale species occur in various taxa including fish, salamanders, insects, shrimp, and spiders [14–17].
One common correlate of pigment level in the integument is environment light level. Pale animals with little to no melanin are commonly active in environments with no ambient light such as caves, the deep sea, soil, and parasites inside the body of hosts [16, 18, 19], but are rare in terrestrial environments. Animals that live in dim light conditions, i.e., active nocturnally, sometimes also have little pigmentation and thus are pale compared to their diurnal congeners [e.g., bees, 20, 21]. However, the adaptive advantage of low pigment levels in low-light environments is unclear given that melanin is relatively cheap to produce and its potential advantages are many [22, 23, but see 24].
Species that deposit little pigment in their integument often display a suite of associated and selectively advantageous traits to low-light environments. One common adaptation in these species, especially those in lightless environments, is a severe reduction in or loss of eyes, with this trait being particularly well-studied in fish and other species that have pigmented terrestrial counterparts with fully developed eyes [14, 15, 25, 26]. Alternatively, many organisms that live in dim light environments and have lost some to most of their pigment possess exceptionally large eyes that enhance visual system performance in low light [20, 21].
Herein, we explore the association between body coloration, daily activity patterns, and eye structure in ant species that vary in the extent of their cuticular pigmentation. We designate two categories of coloration: pale (little pigmentation, appearing mostly whitish-yellow to yellowish to amber), and dark (extensive pigmentation, appearing orange or light to dark brown or black) (see Figs 1–4). Existing knowledge about the relationship between body color, light environment, and eye structure in ants suggests that they display relationships common in other taxa, i.e., (1) compared to close relatives from well-lit environments, species that live in lightless subterranean habitats are paler in color and have eyes that are absent or severely reduced in size [e.g., 27, 28], and (2) pale, nocturnal species that forage above ground have relatively large eyes compared to diurnally foraging species [e.g., 29]. Specifically, this study was motivated by the observation that eyes and facet lenses were larger in pale, nocturnal, above ground foraging species of honey pot ants (Myrmecocystus subgenus Myrmecocytus–subfamily Formicinae) compared to their dark congeners [see 30].
Profile photographs of pale (A-C) and dark species (D-F) (see text) of Myrmecocystus (subfamily Formicinae) examined in this study: (A) M. christineae (CASENT0923358), (B) M. navajo (CASENT0923356), (C) M. mexicanus-02 (CASENT0923355), (D) M. yuma (CASENT0923359), (E) M. kennedyi (CASENT0923362), and (F) M. mendax-03 (ASUSIBR00001132). Note the relatively larger eyes of pale compared to dark species. Species are arranged by size pairs–A&D, B&E, and C&F (see text). Photographs by Michele Esposito from www.antweb.org.
Profile photographs of pale (A) and dark species (B-D) (see text) of Aphaenogaster (subfamily Myrmicinae) examined in this study: (A) A. megommata (CASENT0923367), (B) A. boulderensis (CASENT0005722), (C) A. occidentalis (CASENT0005725), and (D) A. patruelis (CASENT0923366). Note the relatively larger eyes of pale compared to dark species. Photographs by Michele Esposito from www.antweb.org.
Profile photographs of pale (A) and dark species (B–C) (see text) of Temnothorax (subfamily Myrmicinae) examined in this study: (A) T. sp. BCA-5 (CASENT0118165), (B) T. neomexicanus (CASENT0923368), and (C) T. tricarinatus (CASENT0102845). Note the relatively larger eyes of pale compared to dark species. Photographs by Michele Esposito from www.antweb.org.
Profile photographs of pale (A–B) and dark species (C–J) (see text) of Veromessor (subfamily Myrmicinae) examined in this study: (A) V. lariversi (CASENT0923345), (B) V. RAJ-pseu (CASENT0923346), (C) V. andrei (CASENT0923137), (D) V. chamberlini (CASENT0005730), (E) V. chicoensis (CASENT0923347), (F) V. julianus (CASENT0104946), (G) V. lobognathus (CASENT0923126), (H) V. pergandei (CASENT0923124, (I) V. smithi (MCZ-ENT00671466), and (J) V. stoddardi (CASENT0922825). Note the relatively larger eyes of pale compared to dark species. Photographs by Wade Lee, April Nobile, and Michele Esposito from www.antweb.org.
Broadly, we were interested in the evolution and consequences of these associations, and we used a comparative approach to examine these relationships in four ant genera in two subfamilies (Myrmecocystus—subfamily Formicinae and Aphaenogaster, Temnothorax, Veromessor—subfamily Myrmicinae) that contain pale and dark species. This multitaxa approach strengthened our ability to make evolutionary inferences.
This study tests two hypotheses related to body color, activity period, and eye structure. As detailed above, pale species tend to be restricted to lightless and dim light environments. Consequently, we first test the prediction that in four genera of ants activity period of pale species likewise is restricted to dim light conditions, i.e., nocturnal, whereas activity period of their dark congeners is not restricted to dim light conditions. To this end, we first quantified for each species the association between cuticular pigmentation and daily above-ground activity patterns, i.e., whether pale species are more likely to be nocturnal than their dark congeners. Second, we tested the prediction derived from the arguments above that eye structure varies with body color and activity time. Specifically, we examined whether within each genus pale species have larger eyes than their dark congeners, i.e., eyes that would enhance vision in dim conditions. Several studies have compared the compound eyes of nocturnal and diurnal ants [29]. However, most of these comparative studies lacked adequate controls for phylogeny in that they compared relatively small numbers of species from different lineages or with unknown phylogenetic relationships [29, 31, 32]. In this study, phylogenies were available for all four examined ant genera, such that we could compare eye structure of closely related pale and dark species. However, phylogenetic corrections were not performed.
For a subset of these species, we examined eye structure in more detail to explore the effects of activity-pattern-related variation in eye size on visual sensitivity, acuity, and field dimensions. High visual acuity is generally expected for members of the Hymenoptera (i.e., ants, wasps, bees) whose apposition eyes [33] are typically structured to maximize image resolution rather than light capture. Visual resolution of apposition eyes can be assessed by measuring facet diameter (D) and interommatidial angle (Δϕ, the angle between the optical axes of adjacent ommatidia). Larger facets capture more light and D is positively correlated with sensitivity, while resolving power is negatively correlated with Δϕ. Consequently, there may be a potential tradeoff between resolution and sensitivity given that D, Δϕ, and eye size interact in complex and sometimes opposing ways [34]. Nocturnal animals usually resolve this tradeoff in favor of sensitivity, and thus have lower acuity compared to their diurnal counterparts [35].
We also used the eye parameter (ρ, which is the product of D in μm and Δϕ in radians) [35–37], to characterize the compromise between sensitivity and acuity for each species. The calculated ρ indicates how closely the eye develops to the limits imposed by diffraction, i.e., whether the eye is structured to enhance resolution over sensitivity. These values are low for diurnally active species, and they increase at lower light intensities with nocturnal species often having ρ values ≥ 2.
For two species within each genus (one pale, one dark), we also measured Δϕ in the center of the eye, visual field span, and regional variation in D. Collectively, these measures permit inferences about how visual field structure varies between nocturnal and diurnal species both within and across genera.
We also examined variation in size of the ocelli in Myrmecocystus, which was the only examined genus in which workers possess these structures. Ocelli are a second visual system present in most flying insects that detect polarized light and assist in head stabilization and horizon detection [38–40], but also reflect the natural history and environment of the species [41]. Ocelli are present in alate queens and males of nearly all ant species, but they typically are lacking in the pedestrian workers with Myrmecocystus being a notable exception [42, 43]. Snelling [30] noted that the ocelli were smaller in pale than in dark species of Myrmecocystus.
All species examined herein were geographically restricted to the southwestern United States and northwestern Mexico. The relatively large number of pale species found in this region suggests that pale ants may occur in other regions of the world and in other genera and/or subfamilies. Consequently, we assessed variation in ant color and its correlation with eye size on a larger geographical and taxonomic scale by surveying photographs across species in several ant genera. We could not test the correlation between body color and nocturnal activity in these species, but we expect pale body color to be correlated with eye morphology that enhances visual sensitivity.
Methods
Study species
We examined the relationship between body color, activity pattern, and eye structure in 23 ant species spread across four genera in two subfamilies–Myrmecocystus (subfamily Formicinae) and Aphaenogaster, Temnothorax, and Veromessor (subfamily Myrmicinae); all four genera contained pale and dark species. Hereafter, names of pale species are in normal font, and names of dark species in bold font.
Myrmecocystus: We examined 74 workers from six species (Fig 1). This genus is restricted to North America, and it consists of 29 described species [30, 44], plus several undescribed and cryptic species [45]. Species of Myrmecocystus display a wide range of body sizes, and all species are size polymorphic. We compared three size-similar species pairs that differed in pigmentation across this range of body sizes: small (M. christineae and M. yuma), medium (M. navajo and M. kennedyi), and large (M. mexicanus-02 and M. mendax-03). The genus consists of two major clades: most pale species occur in one clade, and the rest of the pale species and all dark species occur in the other clade. The latter clade consists of six smaller clades–one includes the rest of the pale species, and the other five consist of dark species [45]. Consequently, we could not compare eye structure for phylogenetically adjacent pale and dark species. Instead, we chose both pale and dark species pairs based on a combination of availability of specimens, taxonomic stability, and relative ease of identification.
In his revision of Myrmecocystus, Snelling [30] also indicated that the ocelli were reduced in size or absent in pale species compared to their dark congeners. We tested this observation by measuring diameter of the anterior ocellus for workers of the above six species, plus workers of three additional pale species (M. ewarti, M. testaceus, M. mexicanus-01). As such, our analysis included all known pale species except M. melanoticus and M. pyramicus [30, 45], for which specimens were not available.
Aphaenogaster: We examined 38 workers from four species (Fig 2). This genus includes 30 species in North America. We compared A. megommata, the only pale species, with three closely related dark congeners A. boulderensis, A. occidentalis, and A. patruelis [46, 47].
Temnothorax: We examined 29 workers from three species (Fig 3). In North America, this genus consists of more than 80 described species plus numerous undescribed species, with the T. silvestrii clade consisting of several poorly known, poorly collected pale species (M. Prebus, pers. comm.). Our analysis included the undescribed pale species T. sp. BCA-5 [in 48, as Leptothorax sp. BCA-5] from the T. silvestrii group, and T. neomexicanus and T. tricarinatus, which are two dark species from the T. tricarinatus group, which is sister to the T. silvestrii group [49, M. Prebus, pers. comm.].
Veromessor: We examined 133 workers from the 10 species that occur in the genus (Fig 4). This genus only occurs in North America [50, 51]; two species are pale, V. lariversi and V. RAJ-pseu, while the eight other species are dark.
All specimens are deposited in the collections of Robert. A. Johnson, Tempe, AZ (RAJC), and Matthew M. Prebus, Tempe, AZ (MMPC).
Measure of body coloration
All brightness, body size, and eye measurements were made from photographs of workers as described below. Brightness (B) was measured using the color window in Adobe Photoshop from photographs downloaded from Antweb (www.antweb.org). This assessment is similar to the lightness value in HSL that has been used to characterize body color in other studies of ants [10, 12]. We excluded specimens that were obviously discolored, i.e., those in which the color differed substantially from intraspecific specimens recently collected by RAJ. Using the photograph of the worker body in profile, we measured the average B over an 11 × 11 pixel area (ca. 8.2 mm2) on the head (immediately posterior to the eye), mesosoma (center of mesopleura), and gaster (anteroposterior portion of first gastral tergum), then averaged these values for each worker, then averaged that value across all workers for each species. In measuring B, we focused on areas of diffuse reflectance and avoided spots of specular (mirror-like) reflectance so as to minimize their effect on B values. Using the average B value across the three body tagmata accounted for intra-individual variation in color that occurs in many ant species. All measurements were made by RAJ. We compared using a t-test mean B values for species qualitatively categorized as pale versus dark.
Activity patterns
The relationship between color and activity pattern was evaluated by gleaning above-ground foraging times from literature, personal observations, and personal communications. Foraging times were classified as one or more of the following: diurnal, nocturnal, matinal, crepuscular, and variable. Exclusively day-active or night-active species were classified as diurnal and nocturnal, respectively. The category “variable” included species that forage during both day and night and those that have matinal and crepuscular foraging. This includes species in which foraging time varies seasonally with temperature–diurnal when days are cool, crepuscular-matinal as temperatures increase, and nocturnal when nights are warm. We tested the association between color (pale and dark) and activity time (diurnal, nocturnal, variable) using a Fisher’s exact test [52].
Body size and eye measurements
We measured body size and eye characteristics for workers from all 23 species listed above. These measurements included up to 15 workers per species, selected from up to seven colonies (when available), so as to include variation within and across colonies. Workers of several species display weak to strong size polymorphism, and we selected workers within colonies of these species to span the variation in body size. Body size was measured as mesosoma length (= Weber’s length), which is a standard measure for body size in ants [53]. Mesosoma length was measured as the diagonal length of the mesosoma in profile from the point at which the pronotum meets the cervical shield to the posterior base of the metapleural lobe. Mesosoma length was measured from photographs taken using a Spot Insight QE camera attached to a Leica MZ 125 microscope. Images were then displayed on a video monitor, and mesosoma length was measured using ImageJ (available at http://rsb.info.nih.gov/nih-image/). Measurements were calibrated using photographs of an ocular micrometer scaled in 0.01 mm increments.
Eye measurements were made from high-resolution photographs of the left eye taken in profile focused on the center of the eye at an angle that allowed viewing all facets. Photographs were taken using a Leica M205C microscope at 100× that was linked to the stacking software program Helicon Focus (www.heliconsoft.com/heliconsoft-products/helicon-focus). This software combines photographs taken in different focus planes into one photograph where the entire eye surface is in focus. Facet lenses were counted, and eye area and facet diameter (D) were measured using Digimizer (https://www.digimizer.com/). The area tool was used to calculate area. This tool also calculated the centroid of the eye, and D was the average of four adjoining facets at the centroid. We also measured diameter of the anterior ocellus for species of Myrmecocystus. All photographs contained a 0.15 mm scale bar for calibrating measurements.
Detailed eye measurements
Detailed measures of eyes and visual field were taken from a subset of the 23 species. We conducted four eye measurements on two species from each genus, one pale and the other dark, which were closely matched in mesosoma length.
Interommatidial angle (Δϕ), eye parameter (ρ), and visual field.
Measurements of Δϕ allowed us to examine how spatial resolution varied with activity period, eye size, and body size. We measured Δϕ in the lateral region of the eye for five workers from one pale and one dark species in each of the four genera (M. navajo, M. kennedyi in Myrmecocystus; A. megommata, A. occidentalis in Aphaenogaster; T. BCA-5, T. neomexicanus in Temnothorax; V. lariversi, V. chicoensis in Veromessor) using the radius of curvature method outlined in Bergman and Rutowski [54] with minor modifications. For each specimen, we photographed the left eye in side view from a position on a line perpendicular to the anterior-posterior axis of the eye. This created a photograph showing the edge of the eye surface corresponding to the lateral region with individual facets visible at the eye edge. Each image was copied into Geogebra (©International Geogebra Institute, 2013; www.geogebra.org) to measure the angle subtended by the eye surface spanned by two facets in the apical region. Briefly, a point was identified between two facets at the edge of the eye at its apex. We then drew a line to a point on the eye surface two facet rows away in the anterior direction. This was taken as the tangent to the eye at that point and the perpendicular bisector of these lines between these points was drawn. This was also done for a line extending from the original point to a point two facet rows in the posterior direction. The Δϕ was the angle between these two perpendicular bisectors divided by two. We measured Δϕ three times in the same area for each worker and used the average in our analysis. We calculated ρ for each worker as the product of Δϕ in radians and D in μm.
The same images that were used to measure Δϕ also were used to measure the anterior-posterior visual field span. Lines perpendicular to the tangent of the eye surface were drawn at the anterior and posterior edge of each eye. The angular span of the visual field along the anterior-posterior axis was characterized by the angle between these lines. This measurement was repeated three times for each specimen and the mean value was used in our analyses.
Regional variation in D.
Regional variation in D was measured for Myrmecocystus and Veromessor. The small eyes of Aphaenogaster and Temnothorax contained too few facets to warrant examination of regional variation. We quantified regional variation in D using the photographs taken for eye size measurements (see above) for one pale and one dark species of Myrmecocystus (M. navajo and M. kennedyi) and Veromessor (V. lariversi and V. chicocensis). The image for each individual was printed on letter size paper, and D’s were measured in five eye regions: anterior, dorsal, lateral, posterior, and ventral. The anterior-posterior axis of the eye was a line from the mandible to the posterior corner of the head, and eye regions were described relative to this line. In each region, three facets in one row were measured to the nearest 0.1 mm with digital calipers, then scaled using the 0.15 mm scale bar present on each photograph. Mean D in each region was total length divided by three.
Data analysis
Eye area, facet number, and mean D (dependent variables) were compared across species (independent variable) within each genus using a multivariate analysis-of-covariance (MANCOVA) in the general linear models (GLM) program of SPSS [52, 55, 56]; mesosoma length was the covariate. A least significant difference (LSD) post-hoc test compared the estimated marginal means across species for each variable using mesosoma length as a covariate. Diameter of the anterior ocellus (dependent variable) was compared similarly across species (independent variable) of Myrmecocystus using analysis-of-covariance (ANCOVA), with mesosoma length included as the covariate. Myrmecocystus navajo was omitted from this analysis because only one worker had a visible anterior ocellus (see below).
The dependent variables–Δϕ, ρ, and visual field span—were compared in separate ANCOVA’s using genus (4 levels) and activity period (diurnal versus nocturnal) as independent variables, with mesosoma length as a covariate [52]. A Tukey’s HSD post-hoc test compared differences across genera and species for each variable. Regional variation in D was analyzed using a one-way repeated measures ANOVA followed by a LSD post-hoc test for each species [52, 57]. For all tests, data were transformed, as necessary, to meet the assumptions for homogeneity of variance-covariance matrices (Box’s M test and Levene’s test) and homogeneity of regression slopes [56, 58].
Survey for additional pale ant species
We used Antweb (www.antweb.org) to scan photographs for pale ant species in the genera Aphaenogaster, Crematogaster, Messor, and Temnothorax (subfamily Myrmicinae), and Dorymyrmex and Iridomyrmex (subfamily Dolichoderinae). We scrolled through frontal photographs of the head for all species in each genus looking for species that appeared pale. We verified our visual assessment of color for these taxa by measuring their brightness value (B) using Adobe Photoshop, as detailed above, with pale species being those with a mean B value > 70. We then assessed relative eye size for these species, calculated as eye area/mesosoma length, compared to their dark congeners, which all had B values < 70. Phylogenies were not available for these genera, so our dark species comprised the first five species/nominal subspecies on the Antweb page from each genus. Pale species with enlarged eyes were those that had a relative eye size larger than that of all five dark congeners. Photographs of each species can be viewed by going to https://www.antweb.org/advSearch.do, then placing the genus and species name in the advanced search box.
Results
Pigmentation and daily activity pattern
We first quantitatively confirmed our visual impressions of variation in body color. As predicted, our brightness values (B) measured from Antweb photographs were consistently higher for species that we visually classified as pale compared to dark across all four genera (t-test, t = -10.124 df, P < 0.0001). Mean B values were 74.6 (range 65.1–87.8, n = 10) for pale species and 45.4 (range 32.3–58.7, n = 16) for dark species (Table 1). Values did not overlap for any pale and dark species as all dark species had a mean B below 60, whereas all pale species had a mean B above 65. However, note that the two pale species of Veromessor displayed mean B values that ranged from 65–70, which was intermediate to pale and dark species in the other three genera (Table 1). There also was a significant effect of color (pale, dark) on activity period (diurnal, nocturnal, variable) (P < 0.0001, Fisher’s exact test) with a preponderance of pale species that are nocturnal and dark species that have diurnal/variable activity periods (Table 2). Henceforth, we use pale and dark to refer to species with nocturnal and diurnal/variable activity periods, respectively.
Brightness values are given as mean ± 1 SE (number of workers) (see text). Morphological measures are given as mean (± 1 SE) (also see Figs 5 & 7–9); for the three eye traits, values on the first line are raw data, values on second line are estimated marginal mean values using mesosoma length as a covariate in a multivariate analysis-of-covariance. Species are listed alphabetically by subfamily, genus, and species within each genus. Taxonomy follows Bolton [59]; species followed by a number are undescribed or in the process of revision [see 45]. Pale species in normal font; dark species in bold font (see text).
Pale species have a brightness (B) value > 65 and dark species have a B value < 60 as measured in Adobe Photoshop (see text).
Eye area, facet number, and facet diameter
We determined the magnitude of interspecific differences in eye area, facet number, and facet diameter for species in each of the four genera. Before running the MANCOVA for each genus, we tested the homogeneity of variance-covariance (Box’s M test and Levene’s test) and homogeneity of slopes assumptions (Wilks’ lambda for the species × mesosoma interaction effect). Box’s M test for homogeneity of variance-covariance matrices is not robust to violations of this assumption such that an alpha level of P < 0.001 is recommended [69, 70]. Myrmecocystus and Aphaenogaster met both assumptions. Temnothorax met the homogeneity of variance-covariance assumption but not the assumption for homogeneity of regression slopes because the slopes for eye area differed across species (P = 0.007) (Table 3). Consequently, we reran the model after deleting eye area, and the homogeneity of regression slopes assumption was met using the two other dependent variables (facet number, facet diameter) (Table 3). We then analyzed variation in eye area across species separately using a one-way ANOVA. The dependent variable in this model was relative eye size, calculated as eye area/mesosoma length, which standardized eye area for body size (see above). Veromessor did not meet the homogeneity of variance-covariance assumption, so we used Pillai’s trace instead of Wilks’lambda to evaluate multivariate significance [71]. The Levene’s test for Veromessor was also significant for the regression slopes comparison and in the final model. Consequently, we decreased the P value for post-hoc tests to P < 0.01 [70].
The MANCOVA for each genus was run twice; the first run tested homogeneity of variance-covariance and homogeneity of regression slopes assumptions, and the second run was for results of the model. For the Levene’s test, the first column gives the P value for the assumptions run, the second gives the P value for the results run. Values for these three lines are eye area, facet number, and facet diameter, respectively. For Temnothorax, the first line is for all three variables, the second line is the model after deleting eye area (see text); values in the second line for the Levene’s test are for facet number and facet diameter, respectively.
Myrmecocystus.
Eye structure (eye area, facet number, facet diameter) differed among species of Myrmecocystus after controlling for mesosoma length (Wilks’ λ = 0.031, F15,180 = 30.0, P < 0.001). The tests of between-subject effects demonstrated that all three dependent variables differed across species (eye area: F5.67 = 76.4, P < 0.001; facet number: F5,67 = 4.9, P < 0.001; mean D: F5,67 = 127.5, P < 0.001; Fig 5). Based on the estimated marginal means, pairwise comparisons across all species pairs using a LSD test showed that eye area and mean D were larger in pale species (M. christineae, M. navajo, M. mexicanus-02) than in their dark congeners (M. yuma, M. kennedyi, M. mendax-03) (P < 0.001, Fig 5). Facet number varied among species in a different pattern, with this number higher in M. mexicanus-02 and M. yuma, but with M. yuma also overlapping the four other congeners with fewer facets. Relative to the three size-paired pale and dark species (M. christineae vs. M. yuma; M. navajo vs. M. kennedyi; M. mexicanus-02 vs M. mendax-03), mean D (using estimated marginal means) was 1.44× larger for M. christineae (20.81 μm) compared to M. yuma (14.42 μm), 1.43× larger for M. navajo (20.41 μm) compared to M. kennedyi (14.29 μm), and 1.41× larger for M. mexicanus-02 (20.78 μm) compared to M. mendax-03 (14.79 μm) (Table 1).
Eye area (mm2) (A), facet number (B), and mean facet diameter (D) (μm) (C) for species of Myrmecocystus (subfamily Formicinae: tribe Lasiini). Three species are pale (open symbols, normal font, and blue lines: M. christineae, M. navajo, M. mexicanus-02), and three species are dark (filled symbols, bold font, red lines: M. yuma, M. kennedyi, M. mendax-03) (see text). For each species, number of workers examined and number of colonies they were derived from is given in parentheses. Significant differences (P < 0.05) among species are denoted after each species name by the letters a–c: a > b > c; the three sets of letters for each species correspond to panels A, B, and C, respectively. Groupings are based on univariate F tests within MANCOVA using the estimated marginal means followed by pairwise comparisons using a least significant differences test (see text).
Mesosoma length was an important covariate in the model (Wilks’ λ = 0.398, F3,65 = 32.8, P < 0.001), indicating that the covariate adjusted values of the outcome. Tests of between-subjects effects differed for all three variables (eye area: F1,67 = 88.9, P < 0.001; facet number: F1,67 = 95.3, P < 0.001; mean D: F1,67 = 12.8, P = 0.001). All three eye features increased with body size within all six species (Fig 5).
Diameter of the anterior ocellus differed across species after controlling for mesosoma length (ANCOVA: tests of between-subject effects; F7,98 = 69.6, P < 0.001; Fig 6), but the species × mesosoma length interaction term was not significant (F7,91 = 0.96, P = 0.47). Pairwise comparisons between all species pairs using LSD tests showed that anterior ocellus diameter usually was larger for dark species, though the diameter for one pale species (M. christineae) overlapped with this group. Ocellus diameter for the other four pale species (M. testaceus, M. mexicanus-01, M. mexicanus-02, M. navajo) was smaller than all other congeners (Fig 6). Note that M. navajo was not included in our statistical analysis, but it was placed lowest in this group post-hoc because the anterior ocellus was lacking in 11 of 12 workers.
Six species are pale (open or red symbols and normal font: M. christineae, M. ewarti, M. navajo, M. testaceus, M. mexicanus-01, M. mexicanus-02), and three species are dark (filled symbols and bold font: M. yuma, M. kennedyi, M. mendax-03) (see text). For each species, number of workers examined and number of colonies they were derived from is given in parentheses. Significant differences (P < 0.05) among species are denoted after each species name by the letters a–f: a > b > c > d > e > f; the three sets of letters for each species correspond to panels A, B, and C, respectively. Groupings are based on univariate F tests within ANCOVA followed by pairwise comparisons of the estimated marginal means using a least significant differences test (see text).
Mesosoma length was an important covariate in the model based on the tests of between-subjects effects (F1,98 = 171.2, P < 0.001). Diameter of the anterior ocellus increased with body size within all species except M. navajo. Presence of the anterior ocellus also was associated with body size in M. mexicanus-01 and M. mexicanus-02, as workers with a mesosoma length < ≈ 2.2 mm lacked an anterior ocellus while those with a mesosoma length > ≈ 2.2 mm had this ocellus, with ocellus diameter increasing with body size in these latter workers (Fig 6). Both posterior ocelli usually were present, but tiny, in workers that lacked an anterior ocellus.
Aphaenogaster.
Eye structure (eye area, facet number, facet diameter) differed among species of Aphaenogaster after controlling for mesosoma length (MANCOVA: Wilks’ λ = 0.042, F9,76 = 22.6, P < 0.001). Tests of between-subject effects demonstrated that all three dependent variables differed across species (eye area: F3,33 = 125.0, P < 0.001; facet number: F3,33 = 146.8, P < 0.001; mean D: F3,33 = 31.9, P < 0.001; Fig 7). Based on the estimated marginal means, pairwise comparisons across all species pairs using an LSD test showed that all three eye measures were higher for the pale A. megommata than for all three dark congeners (P < 0.001). Mean D (using estimated marginal means) was 1.15× larger for A. megommata (22.78 μm) compared to A. occidentalis (19.87 μm), 1.13× larger than that for A. patruelis (20.22 μm), and 1.23× larger than that for A. boulderensis (18.49 μm) (Table 1).
Eye area (mm2) (A), facet number (B), and mean facet diameter (D) (μm) (C) for species of Aphaenogaster (subfamily Myrmicinae: tribe Stenammini). Aphaenogaster megommata is pale (open symbols and regular font), while A. boulderensis, A. occidentalis, and A. patruelis are dark (filled symbols and bold font) (see text). For each species, number of workers examined and number of colonies they were derived from is given in parentheses. Significant differences (P < 0.05) among species are denoted after each species name by the letters a–c: a > b > c; the three sets of letters for each species correspond to panels A, B, and C, respectively. Groupings are based on univariate F tests within MANCOVA using the estimated marginal means followed by pairwise comparisons using a least significant differences test (see text).
Mesosoma length was an important covariate in the model (Wilks’ λ = 0.502, F3,31 = 10.2, P < 0.001), and tests of between-subjects effects differed for eye area (F1,33 = 28.1, P < 0.001) and facet number (F1,33 = 17.1, P < 0.001), but not for mean D (F1,33 = 2.7, P = 0.11). These patterns were evidenced in that eye area and facet number increased with body size within all three species (Fig 7; A. boulderensis excluded because of small sample size), while mean D increased with body size for A. megommata and A. patruelis, but it decreased with body size for A. occidentalis (Fig 7).
Temnothorax.
Eye structure (facet number, facet diameter) differed among species of Temnothorax after controlling for mesosoma length (MANCOVA: Wilks’ λ = 0.054, F4,48 = 39.5, P < 0.001). The tests of between-subject effects demonstrated that both facet number and facet diameter differed across species (facet number: F2,25 = 20.8, P < 0.001; mean D: F2,25 = 53.4, P < 0.001; Fig 8). Based on estimated marginal means, pairwise comparisons across all species pairs using a LSD test showed that mean D was larger for the pale T. sp. BCA-5 than for the two dark congeners, and that facet number was higher in T. tricarinatus than for both T. neomexicanus and T. sp. BCA-5 (Fig 8). Relative eye area also differed across species (one-way ANOVA: F2,26 = 44.0, P < 0.001), with size being largest for T. sp. BCA-05, intermediate for T. tricarinatus, and smallest for T. neomexicanus (LSD test, P < 0.05; Fig 8). Mean D (using estimated marginal means) was 1.23× larger for T. sp. BCA-5 (17.76 μm) compared to T. neomexicanus (14.43 μm) and 1.32× larger than that for T. tricarinatus (13.42 μm) (Table 1).
Eye area (mm2) (A), facet number (B), and mean facet diameter (D) (μm) (C) for species of Temnothorax (subfamily Myrmicinae: tribe Crematogastrini). Temnothorax sp. BCA-5 is pale (open symbols and regular font), while T. neomexicanus and T. triacarinatus are dark (filled symbols and bold font) (see text). For each species, number of workers examined and number of colonies they were derived from is given in parentheses. Significant differences (P < 0.05) among species are denoted after each species name by the letters a–c: a > b > c; the three sets of letters for each species correspond to panels A, B, and C, respectively. Groupings for facet number and facet diameter are based on univariate F tests within MANCOVA using the estimated marginal means followed by pairwise comparisons using a least significant differences test (see text); groupings for eye area are based on a one-way ANOVA (see text).
Mesosoma length was an important covariate in the model (Wilks’ λ = 0.546, F2,24 = 9.97, P < 0.001), and tests of between-subjects effects differed for facet number (F1,25 = 20.8, P < 0.001), but not for mean D (F1,25 = 3.9, P = 0.058). Eye area and facet number increased with body size within all three species, while mean D increased with body size for T. sp. BCA-5 and T. tricarinatus, but it decreased with body size for T. neomexicanus (Fig 8).
Veromessor.
Eye structure (eye area, facet number, facet diameter) differed among species of Veromessor (MANCOVA: Wilks’ λ = 0.024, F27,351 = 33.9, P < 0.001). The tests of between-subject effects demonstrated that all three dependent variables differed across species (eye area: F9,122 = 117.6, P < 0.001; facet number: F9,122 = 124.4, P < 0.001; mean D: F9,122 = 12.5, P < 0.001; Fig 9). Based on the estimated marginal means, pairwise comparisons across all species pairs using a LSD test showed that eye area was larger for the pale V. RAJ-pseu than for the dark V. smithi, and eyes were larger for both species than for the pale V. lariversi. Eye area was smaller for all other dark congeners (P < 0.01; Fig 9). Facet number was higher for V. RAJ-pseu than for V. smithi and V. lariversi, and facet number for these three species was higher than for all other dark congeners. Mean D (using estimated marginal means) was highest for V. smithi and V. RAJ-pseu, followed by V. lariversi and V. julianus, with the two latter species overlapping with V. RAJ-pseu but not V. smithi. Mean D was lower for all other dark congeners (P < 0.01, Fig 9; Table 1).
Eye area (mm2) (A), facet number (B), and mean facet diameter (D) (μm) (C) for species of Veromessor (subfamily Myrmicinae: tribe Stenammini). Veromessor lariversi and V. RAJ-pseu are pale (open symbols and regular font), while the other eight species are dark (filled symbols and bold font) (see text). For each species, number of workers examined and number of colonies they derived from is given in parentheses. Significant differences (P < 0.01) among species are denoted after each species name by the letters a–g: a > b > c > d > e > f > g; the three sets of letters for each species correspond to panels A, B, and C, respectively. Groupings are based on univariate F tests within MANCOVA using the estimated marginal means followed by pairwise comparisons using a least significant differences test (see text).
Mesosoma length was an important covariate in the model (Wilks’ λ = 0.200, F3,120 = 160.2, P < 0.001), and tests of between-subjects effects differed for all three variables (eye area: F1,122 = 486.0, P < 0.001; facet number: F1,122 = 254.2, P < 0.001; mean D: F1,122 = 40.3, P < 0.001). Eye area and facet number increased with body size within all 10 species of Veromessor, and mean D increased for all species except V. smithi and V. chamberlini (Fig 9).
Detailed eye measurements
Variation in interommatidial angle (Δϕ).
Genus has an important effect on Δϕ (ANCOVA: F3,32 = 10.1, P < 0.001). Although the isolated effect of activity period over Δϕ showed a weak effect (F1,32 = 4.0, P = 0.055), the interaction of genus × activity period showed a strong effect (F3,32 = 7.3, P < 0.001), which indicates that the four genera differed in the direction and magnitude of differences in Δϕ. Across all species, values of Δϕ ranged from 3.5–7° (Fig 10).
Interommatidial angle (Δϕ) for one pale (open circles and regular font) and one dark (filled circles and bold font) (see text) species in each of four ant genera: (A) Myrmecocystus, (B) Aphaenogaster, (C) Temnothorax, and (D) Veromessor. All plots have the same x-axis and y-axis scaling in order to visualize differences between light and dark species across genera. Mean Δϕ (in degrees) is given after each species name with an asterisk denoting the species with a significant larger Δϕ based on a t-test (P < 0.05). The significant interaction of genus × activity period is shown by larger Δϕ’s for pale species of Myrmecocystus and Veromessor, whereas Δϕ was larger for dark species of Aphaenogaster and Temnothorax. Sample size is n = 5 for each species.
Pale species had larger mean Δϕ’s in Myrmecocystus (t-test: t 8 = -3.4, P = 0.01) and Veromessor (t 6 = -0.3, P = 0.77), but dark species had larger Δϕ’s in Aphaenogaster (t 8 = -3.6, P = 0.007) and Temnothorax (t 8 = 2.7, P = 0.027) (Fig 10). Across genera, Δϕ was lowest for Myrmecocystus and Veromessor and highest for Aphaenogaster and Temnothorax (Tukey’s HSD test, P < 0.05) (Fig 10).
Mesosoma length was an important covariate in the above model (F1,31 = 5.4, P = 0.026), a fact largely influenced by Δϕ decreasing in larger workers of Temnothorax and Veromessor, and not varying with mesosoma length within species of Myrmecocystus and Aphaenogaster (Fig 10).
Eye parameter (ρ).
The ANCOVA for ρ was significant for genus (F3,32 = 11.6, P < 0.001), activity period (F1,32 = 11.2, P = 0.002), and the interaction of genus × activity period (F3,32 = 13.1, P < 0.001). As expected, overall ρ was higher for pale (mean = 1.70) than for dark species (mean = 1.51), however, a significant genus × activity period interaction indicated differences in direction and magnitude of these differences (Fig 11). Pale species had the larger mean ρ in Myrmecocystus (t-test: t8 df = -8.9, P < 0.001), while values did not differ between pale and dark species in the other three genera (Veromessor: t8 = -0.3, P = 0.74; Temnothorax: t4 = -1.7, P = 0.16; Aphaenogaster: t8 = -2.2, P = 0.058). Across genera ρ was highest for Aphaenogaster, intermediate for Temnothorax and Veromessor, and lowest for Myrmecocystus (Tukey’s HSD test, P < 0.05; Fig 11).
Eye parameter (ρ) for one pale (open circles and regular font) and one dark (filled circles and bold font) (see text) species in each of four ant genera: (A) Myrmecocystus, (B) Aphaenogaster, (C) Temnothorax, and (D) Veromessor. All plots have the same x-axis and y-axis scaling in order to visualize differences between light and dark species across genera. Mean p is given after each species name with an asterisk denoting the species with a significant larger p based on a t-test (P < 0.05). The significant interaction of genus × activity period is shown by larger differences between light-colored and dark-colored species of Aphaenogaster compared to those in the other three genera. Sample size is n = 5 for each species.
Mesosoma length was not an important covariate in the model (F1,31 = 1.6, P > 0.20). The ρ was not positively or negatively correlated with body size for any of the examined species (Fig 11).
Visual field span.
The ANCOVA for visual field span was significant for genus (F3,32 = 53.6, P < 0.001), activity period (F1,32 = 151.7, P < 0.001), and the interaction of genus × activity period (F3,32 = 14.8, P < 0.001). Visual field span was greater for pale (mean = 98.8o) than for dark species (mean = 73.0o), and the significant genus × activity period interaction indicated that differences between the visual field of pale and dark species were larger in some genera, e.g., Aphaenogaster, than others (Fig 12). Though not always significantly different, pale species had a larger mean visual field in all four genera (Myrmecocystus t-test t5 df = -1.90, P = 0.10; Aphaenogaster: t8 = 12.7, P < 0.001; Temnothorax: t-test t8 = -4.6, P = 0.002; Veromessor: t7 = -6.9, P < 0.001). Across genera the visual field was greatest for Myrmecocystus, intermediate for Aphaenogaster, and smallest in Temnothorax and Veromessor (Tukey’s HSD test, P < 0.05; Fig 12).
Anterior-posterior visual field span (in degrees) for one pale (open circles and regular font) and one dark (filled circles and bold font) (see text) species in each of four ant genera: (A) Myrmecocystus, (B) Aphaenogaster, (C) Temnothorax, and (D) Veromessor. All plots have the same x-axis and y-axis scaling in order to visualize differences between pale and dark species across genera. Mean visual field span (in degrees) is given after each species name with an asterisk denoting the species with a significant larger visual field based on a t-test (P < 0.05); the double asterisk denotes that the t-test was not significant, but that the visual field was significantly larger when including mesosoma length as a covariate. The significant interaction of genus × activity period is shown by larger differences between pale and dark species of Aphaenogaster compared to those in the other three genera. Sample size is n = 5 for each species.
Mesosoma length was not an important covariate in the above model (F1,31 = 2.6, P = 0.11), in part, because of the differing patterns exhibited across species. For example, visual field span was positively correlated with mesosoma length in A. megommata, A. occidentalis, T. BCA-5, T. neomexicanus, and V. lariversi, but these two variables were negatively correlated in M. kennedyi, M. navajo, and V. chicoensis (Fig 12).
The maximum visual span for pale species was 128o for M. navajo and 121o for A. megommata; the maximum for all other species was < 120o. Because their eyes are located on the side of the head, the center of these relatively small visual field spans was directed laterally. This means that, in the ants examined here, there was no forward part of the visual field for either eye directed toward the mouthparts, and so there was no anterior region of binocular vision.
Regional variation in D.
D varied regionally in three (M. navajo, V. chicoensis, V. lariversi) of the four species; all three species met the assumption of sphericity (Table 4; Fig 13). Based on a post-hoc LSD test, the anterior and ventral facets were largest in M. navajo, ventral facets were largest in V. chicoensis, and lateral and ventral facets were largest in V. lariversi. D did not vary across regions in M. kennedyi, but the ventral facets had the largest mean D (Table 4; Fig 13).
Regional variation in facet diameter for one pale (open circles and regular font) and one dark (filled circles and bold font) (see text) species in each of two ant genera: (A) Myrmecocystus and (B) Veromessor. Significant differences within each species are denoted by the letters a–c: a > b > c for pale species; d–f: d > e > f for dark species. Groupings for each species are based on a repeated-measures ANOVA followed by a least significant differences test. Sample size is given after each species name.
Additional pale ant species with enlarged eyes.
We found numerous additional pale ant species with enlarged eyes on Antweb (Table 5 and S1 Table). The combination of pale color and enlarged eyes occurred in numerous additional species of Temnothorax from both the Old and New World, as well as in two additional genera–Dorymyrmex and Iridomyrmex (subfamily Dolichoderinae), which comprises a third subfamily containing pale species. In Temnothorax, this combination of traits evolved in at least two species groups in the United States and Mexico (T. silvestrii and T. tricarinatus) and in at least one species group in northern Africa (T. laurae) (Table 5). Interestingly, some pale species of Temnothorax did not have enlarged eyes, e.g., T. agavicola T. colkendolpheri, and T. indra.
Species are listed alphabetically by subfamily, genus, species group, species, and subspecies. High resolution photographs of each species can be viewed by going to https://www.antweb.org/advSearch.do, then placing the genus and species name in the advanced search box. Brightness values are given as mean (n) (see text). Relative eye size was calculated as eye area/mesosoma length.
Discussion
Foraging and cuticular color
As predicted, worker color was correlated with foraging time across all four genera of ants. Specifically, pale coloration is linked to nocturnal foraging in these and potentially other pale ants. Alternatively, dark species usually forage diurnally, but some species also forage nocturnally during warm seasons, and several species are largely matinal-crepuscular-nocturnal foragers. The above two patterns support the hypothesis that activity pattern can produce strong selection pressure on body coloration and eye structure. These relationships are found in a number of taxa of ants and other organisms [15, 16, 26], but it is not a necessary phenotype given the numerous taxa living in similarly dim light conditions that have retained their pigmentation.
A species-level phylogeny is available for all four genera such that we can infer the direction of trait evolution. These phylogenies indicate that pale color is a derived trait in Aphaenogaster [46], Temnothorax [49], and Veromessor (M. Borowiec & R.A. Johnson, unpub. data), i.e., all most recent common ancestors of pale species were dark, but that it is an ancestral trait in Myrmecocystus [45], i.e., pale color was a basal trait in this genus and that these species gave rise to dark congeners. Van Elst [45] also determined that the subgenus Myrmecocystus (all pale species) and the genus Myrmecocytus as a whole most likely originated from a nocturnal ancestor. Interestingly, the sister genus, Lasius [45], contains numerous pale species that are largely subterranean with very small eyes [72, 73].
Compound eye morphology
We hypothesized that the eyes of pale ant species maximized sensitivity at the expense of resolution, which is the pattern typical for most insects with apposition eyes [35, 74]. The predicted features to enhance sensitivity include larger eye size, larger facets, larger Δϕ‘s, and a higher ρ. Using pale body color as an indicator of nocturnal activity in ants demonstrated consistent correlated adaptations in eye structure across four genera of ants in two subfamilies. When controlled for body size, pale species exhibited convergent morphology for some characters, but not for others: all pale species (except V. lariversi) had larger eyes, a larger D, and a larger visual span compared to their dark congeners. Aphaenogaster megommata and V. RAJ-pseu also possessed more eye facets than their dark congeners. Alternatively, Δϕ and ρ displayed variable patterns both within and among genera.
Sensitivity (light gathering potential) of an eye is a function of four variables that effect photon capture–D, rhabdom diameter, rhabdom length, and focal length [see 29]. Facet area [π/4 × D2] is a particularly important indicator in the differences in light capturing ability of the eyes between pale and dark species in each genus [35]. This difference was highest for Myrmecocystus with facet area for pale species about 2.0–2.1-fold higher than for their paired dark species (calculated using estimated marginal means as [π/4 × D2pale]/ [π/4 × D2dark]; see Table 1), about 1.3–1.5-fold higher for A. megommata compared to its dark congeners, and about 1.5–1.7-fold higher for T. sp. BCA-5 compared to its dark congeners. Alternatively, for Veromessor, facet area was highest in the dark V. smithi. The mean difference was about 1.05–1.10× higher for V. smithi than for V. lariversi and V. RAJ-pseu; all other dark congeners had a smaller D. Moreover, our measurements of D documented clear differences in eye structure and sensitivity between pale and dark species. Future studies should examine the other three above variables so as to more completely understand the suite of structural adaptations possessed by eyes of these pale species.
Although not examined, the eyes in pale species of Myrmecocytus were more protruding and dome-shaped compared to the more flattened eyes of their dark congeners (Fig 1). These more bulging, dome-shaped eyes could result in a greater radius of curvature and possibly a greater visual span field, as well as space for more facets within a given eye area.
Our results show variation in the strength of the relationship between activity patterns, body color, and eye structures that suggest other variables, yet to be determined, have come into play during evolution. Interestingly, the two pale sister species of Veromessor, V. RAJ-pseu and V. lariversi, displayed different patterns of eye structure, with V. RAJ-pseu having larger eyes and more facets than V. lariversi (Fig 9). Additionally, eyes of the dark species V. smithi were smaller with fewer facets than V. RAJ-pseu, but they were larger with larger D’s compared to V. lariversi. This may result from the fact that V. smithi is a nocturnally-active dark species. One difference between V. lariversi and V. RAJ-pseu and other pale species examined herein is their more yellowish-amber to yellowish-orange color and lower B value, indicating that they are less pale than pale species in the other three genera (see Table 1; Figs 1–4). Differences in eye structure across genera along with variation across species of Veromessor, (e.g., V. lariversi, V. RAJ-pseu, and V. smithi) are similar to the wide variation in degree of pigment loss and eye degeneration found among cave-dwelling species that is caused by differences in divergence time and intensity of selection [18]. Similar variation occurs for nocturnal foraging bees in which many species have a relatively pale body color, and many but not all species have enlarged compound eyes and ocelli [20].
Interommatidial angles (Δϕ’s) determine the distance at which objects can be resolved, with resolution distance increasing as Δϕ decreases. All of our species had relatively large Δϕ’s that ranged from 3.5–7o, especially when compared to many common flying insects that have Δϕ’s that range from 1–3o [37]. Pale and dark species varied in their patterns of Δϕ which were significantly larger in the dark Temnothorax and Aphaenogaster, which had small eyes with fewer facets, but was larger for the pale Myrmecocystus which had numerous eye facets. Moreover, Δϕ did not decrease for pale species indicating that daily activity patterns have had little effect on the evolution of resolving power.
The eye parameter (ρ) measures the tradeoff between sensitivity and resolution. Insects active during high light conditions usually have low ρ values that enhance resolution, whereas species active in low light have higher ρ values that often exceed 2 μm rad [36]. Across our four genera, ρ was significantly higher only for the pale M. navajo compared to the dark M. kennedyi, with ρ for the former species approaching 2 (Fig 11). The higher ρ value for M. navajo resulted from the combination of larger facets and a larger Δϕ (Fig 11). In contrast, M. kennedyi was the only strictly diurnal forager among all dark species (Table 1), and correspondingly it had the lowest mean ρ value (0.91) among all species (Fig 11). The ρ value was similar for the other three pairs of congeners, with the dark A. occidentalis having the highest ρ value (2.12) of all species (Fig 11). The lack of consistent patterns across species of Aphaenogaster, Temnothorax and Veromessor likely reflects the wide range of light conditions under which dark species forage including nocturnal foraging in some seasons (Table 1).
The visual field was larger for pale species in all four genera (Fig 12). Moreover, there was no indication that these species had binocular vision in the anterior-posterior direction, that is, they cannot use their eyes for binocular depth perception. This infers that these ants do not use vision to find or capture food items, which aligns with diets that include stationary objects such as seeds, dead insects, and extrafloral nectaries. Instead, it seems likely their eyes are used for detection and orientation relative to land-based and celestial cues used in navigation (see below). In addition, our finding that visual field usually correlated with body size (positively or negatively, depending on the species), contrasted with the pattern for Cataglyphis bicolor, in which visual field was independent of body size [75].
Regional variation in D is common in insects [76], with these size differences probably related to the different selection pressures on eye structure in each region. Larger facets imply that insects have better vision from regions containing larger facets. In this study, ventral facets were significantly larger in three of the four examined species (along with anterior facets in two species) (Fig 13). This general pattern suggests that ventral facets are important for vision in both diurnal and nocturnal activity, perhaps as a mechanism for optic flow to measure distance [see 77, 78].
Other pale ants with enlarged eyes.
Our survey of images on Antweb revealed the occurrence of pale ants with enlarged eyes compared to their dark congeners not just in the clades studied here but also in other genera. Moreover, these coupled traits appear to have evolved independently multiple times across at least three subfamilies. Given the limited scope of museum specimens available, it is likely that many pale species remain to be discovered. As these are located, our technique for measuring brightness provides a tool for mapping patterns of pigment loss within and across ant genera.
One commonality among pale species examined herein and in Table 5 is that many of these species are largely restricted to desert and semi-arid habitats. As such, these species possess visual adaptations to be nocturnal specialists in extreme environments in a manner similar to heat tolerance adaptations possessed by their thermophilic diurnal counterparts such as Myrmecocytus kennedyi and Forelius spp. [30; R.A. Johnson, pers. obs.] in the New World, and Cataglyphis spp. and Melophorus bagoti in the Old World [79, 80]. The open, exposed nature of their foraging environment lacks overstory which suggests that these species can obtain navigation cues from local landmarks via their enlarged eye facets, but probably only horizon and lunar night sky cues. However, at this point, nothing is known about navigation in any pale species, and among dark species, orientation and navigation have been examined only in the column-foraging and mostly diurnal V. pergandei [81–83]. There is much to learn about how ants use their eyes both at night and during the day.
Ocelli. Size of the anterior ocellus varied among pale and dark species of Mymecocystus. In larger species, the anterior ocellus was smaller in pale compared to dark species, but this difference largely disappeared for smaller species (Fig 6). The two largest pale species (M. mexicanus-01, M. mexicanus-02) also displayed size-dependent presence of the anterior ocellus as it was present only in larger workers. The anterior ocellus also was absent in nearly all workers of the intermediate sized M. navajo. The pattern was mixed for smaller species because the anterior ocellus was largest for the dark M. yuma, intermediate for the dark M. kennedyi and pale M. ewarti, and smallest for the pale M. testaceus (Fig 6). These results contrast with other social insects in which the anterior ocellus is typically larger in nocturnal compared to crepuscular and diurnal flying species, e.g., bees and ants [33, 84–87] and pedestrian workers in the ant genus Myrmecia [88].
Absence of the anterior ocellus in workers of some pale species displays a phylogenetic component. Pale species in which some to most workers lacked the anterior ocellus fell into one clade, while all other species that always have an anterior ocellus were in two other clades [see 45]. We were unable to examine the two other pale species (M. pyramicus, M. melanoticus) because specimens were unavailable, but this phylogenetic association predicts that the anterior ocellus always is present in M. pyramicus and that it is only present in larger workers of M. melanoticus (see Fig 6). To our knowledge, these are the only known ant species in which workers display intraspecific variation in presence-absence of the anterior ocellus, making them excellent candidates to examine evolution, development, and function of the anterior ocellus, as well as how such variation affects forager orientation and navigation (see below). Foraging behavior is poorly documented in pale species, but it appears that both small and large workers of M. mexicanus-02, i.e., those with and without an anterior ocellus, leave the nest to forage (J. Conway, pers. comm.).
The function of ocelli in ant workers is poorly understood because most species lack ocelli (notable exceptions include the genera Cataglyphis, Formica, Myrmecocystus, Polyergus in the subfamily Formicinae; Myrmecia in the subfamily Myrmeciinae; Pseudomyrmex in the subfamily Pseudomyrmecinae). Flying insects have three ocelli that serve the general purpose of sensing polarized light for navigation and maintaining flight stability, whereas worker ants that have been studied use their ocelli to detect polarized light for navigation in Cataglyphis bicolor [89] or to gather light in Myrmecia [88]. Ocelli almost always are present in the flying queen and male ants, such that species in which the pedestrian workers possess ocelli provide an avenue to compare ocellar structure and function relative to mode of locomotion and caste. The only such comparison using the nocturnal Myrmecia nigriceps found structural differences between workers and males that included presence of an ocellar tapetum in males but not in workers [88].
Lastly, compound eyes and ocelli provide separate and functionally different visual pathways, so it is instructive to test for convergence in the two pathways. Two studies compared compound eyes and ocelli between nocturnal and diurnal ant species. In leafcutter ants (genus Atta), both the ocelli and eye facets were larger in nocturnal compared to diurnal species of both flying queens and males, while eye area was similar for species in both activity groups [85]. The other study examined workers of four species of Myrmecia also finding that both the ocelli and eye facets were larger in nocturnal compared to diurnal congeners, while number of facets was similar for species in both activity groups [29, 88]. Alternatively, we report here that pale species of Myrmecocystus had an anterior ocellus that was similar in size to smaller than comparable dark congeners, but that eye facet diameter and eye size were larger for pale compared to dark species. However, like the pattern found in Myrmecia [88], facet number was similar for species of Myrmecocystus in both activity groups.
Conclusions
This study provides a first overview of variation in external eye structure across several ant genera that compares closely related pale and dark congeners. Our observations on body coloration and eye structure allow several statements about their visual ecology. First, the correlation between ant body color and activity period parallels that found in other animals. The specific selective factors shaping this correlation await more detailed work on the costs and benefits of cuticular pigmentation. Second, pale, above ground foraging ants have enlarged rather than reduced or no eyes relative to their dark congeners, suggesting that vision is important for both nocturnal and diurnal species across several lineages. That pale species possess optical adaptations to maximize sensitivity, which is the pattern typical for most nocturnal insects with apposition eyes [35, 74], also suggests that vision plays a role in navigation for these nocturnal ants. Third, the visual field span and mild regional variation in D suggest that their eyes are not adapted to gather detailed visual information from any specific region in the space around the ant, but rather they are gathering relatively low quality information from a large part of the space around them. Fourth, the mild differences in eye structure between pale and dark species suggest both groups use their eyes in similar ways, and they are consistent with observations that these ants use their vision in navigation guided by large celestial and landmark cues. Field studies that detail foraging behavior and navigational skills would complement these data. Additional research should be done to more thoroughly determine optical sensitivity. This study only examined facet diameter, but data are needed on rhabdom diameter, rhabdom length, focal length, and neural adaptations to more completely determine and compare optical sensitivity [see 29]. We also note that activity period is the primary difference between our pale and dark species given that life history and behavior are similar for species within each genus, i.e., most species have solitary foragers that harvest seeds, or scavenge for debris, dead insects, and plant exudates [30, 50: R.A. Johnson, pers. obs.], probably using olfactory and/or tactile cues.
Of the genera examined herein, we believe Myrmecocystus has the most potential for further study given the consistently large variation in eye structure between pale and dark species (eye area, D, ρ, visual span), combined with the fact that most species are strongly polymorphic such that traits can be compared allometrically [see 75]. Additionally, this is the only known genus with pale species that possess ocelli, such that it provides an excellent group to examine internal eye structure and to compare evolution of both eyes and ocelli. The flying queens and males might also be examined for comparative study of the sexual castes, especially given that the queen of M. navajo has extremely large ocelli.
Supporting information
S1 Table. Brightness values and relative eye size for dark congeners from the ant genera Dorymyrmex, Iridomyrmex, and Temnothorax.
Phylogenies were not available for these genera, so we analyzed the first five species/nominal subspecies from each genus on Antweb (www.antweb.org). Species are listed alphabetically by subfamily, genus, species group, species, and subspecies. High resolution photographs of each species can be viewed at https://www.antweb.org/advSearch.do, then typing in the genus and species name in the advanced search box. Brightness values are given as mean (n) (see text). Relative eye size was calculated as eye area/mesosoma length.
https://doi.org/10.1371/journal.pone.0257779.s001
(DOCX)
Acknowledgments
We thank Brian Fisher, Michele Esposito, and Antweb for the high resolution photographs, Christian Rabeling for use of his microscope and photographic programs, and Matt Prebus for the loan of specimens. We also thank two anonymous reviewers for comments that improved this manuscript.
References
- 1. True JR. Insect melanism: the molecules matter. Trends Ecol Evol. 2003;18:640–7.
- 2. Shawkey MD, D’Alba L. Interactions between colour-producing mechanisms and their effects on the integumentary colour palette. Philos T R Soc B. 2017;372:20160536. pmid:28533449
- 3. Hines HM, Witkowski P, Wilson JS, Wakamatsu K. Melanic variation underlies aposematic color variation in two hymenopteran mimicry systems. PLoS One. 2017;12:e0182135. pmid:28753659
- 4. Malcicka M, Bezemer TM, Visser B, Bloemberg M, P. CJ, Snart CJP, et al. Multi-trait mimicry of ants by a parasitoid wasp. Sci Rep. 2015;5:8043. pmid:25622726
- 5. Wellenreuther M. Balancing selection maintains cryptic colour morphs. Mol Ecol. 2017;26:6185–8. pmid:29206344
- 6. Wiens JJ, Tuschhoff E. Songs versus colours versus horns: what explains the diversity of sexually selected traits? Biol Rev. 2020;95:847–64. pmid:32092241
- 7. Dubovskiy JM, Whitten MMA, Kryukov VY, Yaroslavtseva ON, Grizanova EV, Greig C, et al. More than a colour change: insect melanism, disease resistance and fecundity. Proc R Soc Lond Ser B. 2013;20130584. pmid:23698007
- 8. Badejo O, Skaldina O, Gilev A, Sorvari J. Benefits of insect colours: a review from social insect studies. Oecologia. 2020;194:27–40. pmid:32876763
- 9. de Souza AR, Mayorquin AZ, Sarmiento CE. Paper wasps are darker at high elevation. J Therm Biol. 2020;89:102535. pmid:32364980
- 10. Law SJ, Bishop TR, Eggleton P, Griggiths H, Ashton L, Parr C. Darker ants dominate the canopy: testing macroecological hypotheses for patterns in colour along a microclimatic gradient. J Anim Ecol. 2020;89:348–59. pmid:31637702
- 11. Forsman A, Ahnesjo J, Caesar S, Karlsson N. A model of ecological and evolutionary consequences of color polymorphism. Ecology. 2008;90:34–40.
- 12. Bishop TR, Robertson MP, Gibb H, van Rensburg BJ, Braschler B, Chown SL, et al. Ant assemblages have darker and larger members in cold environments. Global Ecol Biogeogr. 2016;25:1489–99.
- 13. Pinkert S, Brandl R, Zeuss D. Colour lightness of dragonfly assemblages across North America and Europe. Ecography. 2017;40:1110–7.
- 14. Protas M, Jefferey WR. Evolution and development in cave animals: from fish to crustaceans. Wiley Interdiscip Rev Dev Biol. 2012;1:823–45. pmid:23580903
- 15. Protas ME, Trontelj P, Patel NH. Genetic basis of eye and pigment loss in the cave crustacean, Asellus aquaticus. Proc Natl Acad Sci USA. 2011;108:5702–7.
- 16. Bilandžija H, Ćetković H, Jefferey WR. Evolution of albinism in cave planthoppers by a convergent defect in the first step of melanin biosynthesis. Evol Dev. 2012;14:196–203. pmid:23017027
- 17. Lohaj R, Casale A. Laemostenus (Iranosphodrus) rudichae, new subgenus and new species of sphodrine beetle from Iran (Coleoptera: Carabidae: Sphodrini). Acta Entomol Slov. 2011;19:43–50.
- 18.
Culver DC. Cave life–evolution and ecology. Cambridge, Massachusetts: Harvard University Press; 1982.
- 19. Li C, Chen H, Zhao Y, Chen S, Xiao H. Comparative transcriptomics reveals the molecular genetic basis of pigmentation loss in Sinocyclocheilus cavefishes. Ecol Evol. 2020;10:14256–71.
- 20. Wcislo WT, Tierney SM. Behavioural environments and niche construction: the evolution of dim-light foraging in bees. Biol Rev. 2009;84:19–37. pmid:19046401
- 21. Engel MS. Classification of the bee tribe Augochlorini (Hymenoptera: Halictidae). Bull Am Mus Nat Hist. 2000;250:1–90.
- 22. Oxford GS, Gilliespie RG. Evolution and ecology of spider coloration. Annu Rev Entomol. 1998;43:619–43. pmid:15012400
- 23. Poulson TL. Evolutionary reduction by neutral mutations: plausibility arguments and data from amblyopsid fishes and linyphiid spiders. Bull Natl Speleol Soc. 1985;47:109–17.
- 24. Bilandžija H, Abraham L, Ma L, Renner KJ, Jeffery WR. Behavioural changes controlled by catecholaminergic systems explain recurrent loss of pigmentation in cavefish. Proc Roy Soc B. 2018;285:20180243. pmid:29720416
- 25. Protas M, Tabansky I, Conrad M, Gross JB, Vidal O, Tabin CJ, et al. Multi-trait evolution in a cave fish, Astyanax mexicanus. Evol Dev. 2008;10:196–209.
- 26. Re C, Fišer Ž, Perez J, Tacdol A, Trontelj P, Protas M. Common genetic basis of eye and pigment loss in two distinct cave populations of the isopod crustacean Asellus aquaticus. Integrat Comp Biol. 2018;58:421–30.
- 27. Rabeling C, Brown JM, Verhaagh M. Newly discovered sister lineage shed light on early ant evolution. Proc Natl Acad Sci USA. 2008;105:14913–7.
- 28. Hosoishi S, Yamane S, Ogata K. Subterranean species of the ant genus Crematogaster in Asia (Hymenoptera: Formicidae). Entomol Sci. 2010;13:345–50.
- 29. Narendra A, Kamhi JF, Ogawa Y. Moving in dim light: behavioral and visual adapations in nocturnal ants. Integrat Comp Biol. 2017;57:1104–16.
- 30. Snelling RR. A revision of the honey ants, genus Myrmecocystus (Hymenoptera: Formicidae). Nat Hist Mus Los Angeles Co Bull. 1976;24:1–163.
- 31. Menzi U. Visual adaptation in nocturnal and diurnal ants. J Comp Physiol A. 1987;160:11–21.
- 32. Yilmaz A, Aksoy V, Camlitepe Y, Giurfa M. Eye structure, activity rhythms, and visually-driven behavior are tuned to visual niche in ants. Front Behav Neurosci. 2014;8:205. pmid:24982621
- 33. Warrant EJ, Kelber A, Wallén R, Wcislo WT. Ocellar optics in nocturnal and diurnal bees and wasps. Arthropod Struc Dev. 2006;35:293–305. pmid:18089077
- 34. Rutowski RL, Gislén L, Warrant EJ. Visual acuity and sensitivity increase allometrically with body size in butterflies. Arthropod Struc Dev. 2009;38:91–100. pmid:18809509
- 35. Greiner B. Adaptations for nocturnal vision in insect apposition eyes. Int Rev Cytol. 2006;250:1–46. pmid:16861062
- 36.
Snyder AW. Physics of vision in compound eyes. In: Autrum H, editor. Handbook of sensory physiology, Vol VII/6A. Berlin: Springer; 1979. pp. 225–313.
- 37. Land MF. Visual acuity in insects. Annu Rev Entomol. 1997;42:147–77. pmid:15012311
- 38. Wilson M. The functional organisation of locust ocelli. J Comp Physiol A. 1978;124:297–316.
- 39. Mizunami M. Information processing in the insect ocellar system: comparative approaches to the evolution of visual processing and neural circuits. Adv Insect Physiol. 1994;25:151–265.
- 40. Simmons PJ. Signal processing in a simple visual system: the locust ocellar system and its synapses. Microsc Res Tech. 2002;56:270–80. pmid:11877802
- 41. Berry RP, Wcislo WT, Warrant EJ. Ocellar adaptations for dim light vision in a nocturnal bee. J Exp Biol. 2011;214:1283–93. pmid:21430205
- 42.
Zeil J, Ribi WA, Narendra A. Polarisation vision in ants, bees and wasps. In: Horváth G, editor. Polarized light and polarization vision in animal aciences. Berlin: Springer-Verlag; 2014. pp. 41–60.
- 43.
Bolton B. Identification guide to the ant genera of the world. Cambridge, Massachusetts: Harvard University Press; 1994.
- 44. Snelling RR. A revision of the honey ants, genus Myrmecocystus, first supplement (Hymenoptera: Formicidae). Bull Southern Cal Acad Sci. 1982;81:69–86.
- 45. van Elst T, Ericksson TH, Gadau J, Johnson RA, Rabeling C, Taylor JE, et al. Reconstructing the evolutionary history of the honey ant genus Myrmecocystus. Mol Phylogenet Evol. 2021;155:107036.
- 46. DeMarco BB, Cognato AI. A mulitple-gene phylogeny reveals polyphyly among eastern North American Aphaenogaster species (Hymenoptera: Formicidae). Zool Scr. 2016;45:512–20.
- 47. Demarco BB, Cognato AI. Phylogenetic analysis of Aphaenogaster supports the resurrection of Novomessor (Hymenoptera: Formicidae). Ann Entomol Soc Am. 2015;108:201–10.
- 48. Johnson RA, Ward PS. Biogeography and endemism of ants (Hymenoptera: Formicidae) in Baja California, Mexico: a first overview. J Biogeogr. 2002;29:1009–26.
- 49. Prebus M. Insights into the evolution and biogeography and natural history of the acorn ants, genus Temnothorax Mayr (Hymenoptera: Formicidae). BMC Evol Biol. 2017;17:250.
- 50. Johnson RA. Biogeography and community structure of North American seed-harvester ants. Annu Rev Entomol. 2001;46:1–29. pmid:11112162
- 51. Johnson RA. Seed-harvester ants (Hymenoptera: Formicidae) of North America: an overview of ecology and biogeography. Sociobiology. 2000;36:89–122.
- 52.
SPSS. SPSS reference guide. Chicago: SPSS, Inc; 1990.
- 53. Parr CL, Dunn RR, Sanders NJ, Weiser MD, Photakis M, Bishop TR, et al. GlobalAnts: a new database on the geography of ant traits (Hymenoptera: Formicidae). Insect Conserv Divers. 2017;10:5–20.
- 54. Bergman M, Rutowski RL. Eye morphology and visual acuity in the Pipevine Swallowtail (Battus philenor) studied with a new method of measuring interommatidial angles. Biol J Linn Soc. 2016;117:646–54.
- 55.
Dattalo P. Analysis of multiple dependent variables. New York: Oxford University Press; 2013.
- 56.
Leard Statistics. One-way MANCOVA in SPSS statistics. Lund Research Ltd: 2018 [cited 2022 February 18]. Available from: https://statistics.laerd.com/spss-tutorials/one-way-mancova-using-spss-statistics.php.
- 57.
Crowder MJ, Hand DJ. Analysis of repeated measures. New York: Routledge; 2020.
- 58.
Leard Statistics. ANOVA with repeated measures using SPSS statistics. Lund Research Ltd: 2018 [cited 2022 18 February]. Available from: https://statistics.laerd.com/spss-tutorials/one-way-anova-repeated-measures-using-spss-statistics.php.
- 59. Bolton B. An online catalog of the ants of the world. 2020 Available from http://antcat.org, accessed July 15, 2022.
- 60.
Snelling RR, George CD. The taxonomy, distribution and ecology of California desert ants (Hymenoptera: Formicidae). Report to Bureau of Land Management, US Department of Interior, Riverside, California. 335 pp + 89 pp. 1979.
- 61.
DeMarco BB. Phylogeny of North American Aphaenogaster species (Hymenoptera: Formicidae) reconstructed with morphological and DNA data. PhD dissertation: Michigan State University, East Lansing, Michigan, 147 pp; 2015.
- 62. Hobbs RJ. Harvester ant foraging and plant species distribution. Oecologia. 1985;67:519–23.
- 63. Brown MJF, Gordon DM. How resources and encounters affect the distribution of foraging activity in a seed-harvesting ant. Behav Ecol Sociobiol. 2000;47:195–203.
- 64. Hamm CA, Kamansky B. New record of Messor chicoensis (Hymenoptera: Formicidae) from the San Joaquin Valley of California. Sociobiology. 2009;53:543–7.
- 65. Johnson RA. Reproductive biology of the seed-harvester ants Messor julianus (Pergande) and Messor pergandei (Mayr) (Hymenoptera: Formicidae) in Baja California, Mexico. J Hymenopt Res. 2000;9:377–84.
- 66. Wheeler GC, Wheeler J. Veromessor lobognathus: second note (Hymenoptera: Formicidae). Ann Entomol Soc Am. 1959;52:176–9.
- 67. Cole AC. A new species of Veromessor from the Nevada test site and notes on related species (Hymenoptera: Formicidae). Ann Entomol Soc Am. 1963;56:678–82.
- 68. Cole AC. Ants of the Nevada test site. Brigham Young Univ Sci Bull Biol Ser. 1966;7:1–27.
- 69.
Hahs-Vaughn DL. Applied multivariate statistical concepts. New York: Routledge; 2017.
- 70.
Tabachnick BG, Fidell LS. Using multivariate statistics. 6th ed. Upper Saddle River, New Jersey: Pearson; 2014.
- 71. Olson CL. Practical considerations in choosing a MANOVA test statistic: a rejoinder to Stevens. Psychol Bull. 1979;86:1350–2.
- 72. Wing MW. Taxonomic revision of the Nearctic genus Acanthomyops (Hymenoptera: Formicidae). Mem Cornell Univ Agric Exp Sta. 1968;405:1–173.
- 73. Wilson EO. A monographic revision of the ant genus Lasius. Bull Mus Comp Zool. 1955;113:1–201.
- 74. Stöckl A, Smolka J, O’Carroll D, Warrant E. Resolving the trade-off between visual sensitivity and spatial acuity—lessons from hawkmoths. Integrat Comp Biol. 2017;57:1093–103. pmid:28992251
- 75. Zollikofer CPE, Wehner R, Fukushi T. Optical scaling in conspecific Cataglyphis ants. J Exp Biol. 1995;198:1637–46.
- 76. Zagorski ER, Merry JW. How do eye size and facet lens size vary by age and sex in Acheta domesticus? BIOS. 2014;85:151–9.
- 77. Ronacher B, Gallizzi K, Wohlgemuth S, Wehner R. Lateral optic flow does not influence distance estimation in the desert ant Cataglyphis fortis. J Exp Biol. 2000;203:1113–21.
- 78. Ronacher B, Wehner M. Desert ants Cataglyphis fortis use self-induced optic flow to measure distances travelled. J Comp Physiol A. 1995;177:21–7.
- 79. Christian KA, Morton SR. Extreme thermophilia in a central Australian ant, Melophorus bagoti. Physiol Zool. 1992;65:885–905.
- 80. Pfeffer SE, Wahl VL, Wittlinger M, Wolf H. High-speed locomotion in the Saharan silver ant, Cataglyphis bombycina. J Exp Biol. 2019;222:jeb198705.
- 81. Freas CA, Congdon JV, Plowes NJR, Spetch ML. Same but different: socially foraging ants backtrack like individually foraging ants but use different mechanisms. J Insect Physiol. 2019;118:103944. pmid:31520596
- 82. Freas CA, Congdon JV, Plowes NJR, Spetch ML. Pheromone cue triggers switch between vectors in the desert harvest ant, Veromessor pergandei. Anim Cogn. 2020;23:1087–105.
- 83. Freas CA, Plowes NJR, Spetch ML. Not just going with the flow: foraging ants attend to polarised light even while on the pheromone trail. J Comp Physiol A. 2019;205:755–67. pmid:31422422
- 84. Kerfoot WB. Correlation between ocellar size and the foraging activities of bees (Hymenoptera; Apoidea). Am Nat. 1967;101:65–70.
- 85. Moser JC, Reeve JD, Bento JMS, Della Lucia TMC, Cameron RS. Eye size and behaviour of day- and night-flying leafcutting ant alates. J Zool. 2004;264:69–75.
- 86. Somanathan H, Kelber A, Borges RM, Wallén R, Warrant EJ. Visual ecology of Indian carpenter bees II: adaptations of eyes and ocelli to nocturnal and diurnal lifestyles. J Comp Physiol A. 2009;195:571–83. pmid:19363615
- 87. Coody CJ, Watkins JF II. The correlation of eye size with circadian flight periodicity of nearctic army ant males of the genus Neivamyrmex (Hymenoptera; Formicidae, Ecitoninae). Tex J Sci. 1986;38:3–7.
- 88. Narendra A, Ribi WA. Ocellar structure is driven by mode of locomotion and activity time in Myrmecia ants. J Exp Biol. 2017;220:4383–90.
- 89. Fent K, Wehner R. Ocelli: a celestial compass in the desert ant Cataglyphis. Science. 1985;228:192–4.