Vision Research 67 (2012) 28–36

Contents lists available at SciVerse ScienceDirect

Vision Research

journal homepage: www.elsevier.com/locate/visres

Behavioural analysis of chromatic and achromatic vision in the cunicularia (: Formicidae) ⇑ Volkan Aksoy , Yilmaz Camlitepe

Trakya University, Department of Biology, Faculty of Sciences, Balkan Campus, 22030 Edirne, article info abstract

Article history: Responses of Formica cunicularia foragers to monochromatic light stimuli of 370, 440, 540, 590 and Received 21 February 2012 640 nm were evaluated in different experimental conditions using a Y-maze apparatus and a circular ori- Received in revised form 8 May 2012 entation platform. The results showed that foragers responded significantly to all test wavelengths at cer- Available online 28 June 2012 tain intensities but could only discriminate 370 and 540 nm from alternatives irrespective of intensity changes. Furthermore, they were also capable of discriminating two long wavelengths, 590 and Keywords: 640 nm, using a photon catch mechanism by their green photoreceptors. Foragers also discriminated Formicidae stimuli pairs of same wavelengths based only on intensity differences they provide. The overall results Formica cunicularia show that F. cunicularia foragers have a dichromatic colour vision system based on inputs of two possible Dichromacy Achromatic vision photoreceptor types sensitive to UV and green. The results also yielded evidence showing that their visual Intensity discrimination systems provided foragers a sensitivity also for wavelengths corresponding to blue and red ranges of the spectrum. Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction tors whose sensitivity maxima correspond to these values are also used in behavioural context remain to be shown. The wood ant For- colour vision has been a popular field of study for scien- mica polyctena and the desert ant Cataglyphis bicolor are the first tists since the pioneering work of von Frisch (1914) on honeybees ant species for which a true colour vision was proved. Results of who convincingly demonstrated that bees could be trained to dis- electrophysiological and behavioural studies revealed a UV–green criminate red, yellow and green from blue and violet. Later on, dichromatic colour vision in F. polyctena workers (Kiepenheuer, Kühn (1924) reported additional data for honeybee vision with 1968; Menzel, 1973; Menzel & Knaut, 1973; Roth & Menzel, which he showed that bee colour vision extended to UV part of 1972). Although results of electrophysiological and behavioural the spectrum. Since then, studies concerning insect colour vision, studies differed from each other, at least a UV–green dichromatic either behaviourally or with electrophysiological methods, ex- colour vision system was suggested for C. bicolor (Labhart, 1986; tended to members of various insect orders but what we know to- Mote & Wehner, 1980; Wehner & Toggweiler, 1972). Electrophys- day come mainly from those on flower visiting hymenopterans and iological studies on Lasius niger and Myrmecia gulosa also revealed lepidopterans (for a review see Briscoe & Chittka, 2001; Kelber, evidence on presence UV–green photoreceptors (Mazokhin- Vorobyev, & Osorio, 2003; Menzel & Backhaus, 1991). Some ant Porshnyakov & Trenn, 1972; Menzel & Blakers, 1975). In an earlier species were also tested for their responses to spectral stimuli study with Formica cunicularia, Mazokhin-Porshnyakov (1974) but present data on ant colour vision is actually very few. Two ear- reported that the maximum of the spectral sensitivity curve of liest studies reported that reacted strongly to UV light but F. cunicularia was at 510 nm. A sensitivity for 365 nm, as high as they were insensitive to red light (Lubbock, 1929; Turner, 1907). 20% of the peak for 510 nm, was also suggested. On the other hand, Later, Tsuneki (1953) studied responses of Camponotus obscripes using the flickering colourimetry method, he offered two and Leptothorax spinosior workers to spectral stimuli and found receptor types for the same species, one at 470 nm and the other no colour discrimination and reported that discrimination ability at 520 nm (Mazokhin-Porshnyakov, 1974). These earlier results on was rather intensity depended. Marak and Wolken (1965) obtained F. cunicularia remain to be confirmed by behavioural studies. an action spectrum for Solenopsis saevissima represented with Spectral stimuli provide not only chromatic but also three peaks at 360, 505 and 620 nm but whether the photorecep- achromatic, that is intensity related cues, as well. Thus, responses of insects to spectral stimuli can depend on either chromatic or achromatic cues, or may be influenced by both. Insects such as ⇑ Corresponding author. Fax: +90 284 2354010. honeybees and moths respond both to chromatic and achromatic E-mail addresses: [email protected] (V. Aksoy), [email protected] (Y. Camlitepe). aspects of coloured stimuli, but the former is learnt faster and used

0042-6989/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.visres.2012.06.013 V. Aksoy, Y. Camlitepe / Vision Research 67 (2012) 28–36 29 more reliably (Chittka et al., 1992; Kelber, 2005; Labhart, 1974; 2. Methods Menzel & Backhaus, 1991). On the other hand, Skorupski and Chittka (2010) proved with intracellular recordings that the re- 2.1. The ants sponses of green photoreceptors of bumblebee workers, which are involved in achromatic contrast in motion detection, were Live specimens of F. cunicularia were obtained from a nest faster compared to those of UV and blue photoreceptors. As in (41°3804600N/26°3701200E) in the garden of Biology Department of many insects, ants also tend to orient towards a light stimulus Trakya University, Edirne. Several hundred workers were trans- and this tendency increases with increasing intensity of the stimu- ferred to a laboratory with their original nest material. They were lus (Kretz, 1979). Tsuneki (1953) showed that light intensity placed in an arena (750 mm diameter 500 mm high) from which played a major role in orientations of C. obscripes and L. spinosior. escape was prevented with a fluon barrier inside the rim. They Light intensity also influenced the responses of S. saevissima were left free for a week to construct their nest and to get accus- foragers (Marak & Wolken, 1965). C. bicolor foragers were tested tomed to laboratory conditions. The laboratory was uniformly illu- for their abilities in a discrimination task between two light stimuli minated by diffused fluorescent light to provide a 12:12 light–dark of the same spectral compositions (340, 434, 493 and 574 nm) but regime and the temperature was kept at 25 °C. A humidifier (Vapac with different intensities (Kretz, 1979). Foragers could microvap PV4, Edenbridge, Kent, UK) was used to provide a relative discriminate the stimuli but the rate of the correct choices for humidity of 50%. the trained stimulus decreased with decreasing intensity difference between the two stimuli. This study, therefore, aims to obtain behavioural responses of 2.2. Experimental apparatus the black meadow ant F. cunicularia foragers to monochromatic spectral stimuli in order to determine (i) their responses to Colour vision and intensity discrimination experiments were decreasing intensities of different wavelengths, (ii) their discrimi- performed in a Y-maze choice apparatus, whereas intensity re- nation abilities between different wavelengths (colour vision), sponse experiments were performed on a circular orientation and (iii) between pairs of same wavelengths with different platform. intensities. The Y-maze was made of glass (30 mm diameter) with the two arms at 120° (Fig. 1a). The base of the Y (500 mm long) was

Fig. 1. Arrangement of (a) the Y-maze and (b) the orientation platform. In the Y-maze design (a), foragers were trained to associate a food reward with one of two wavelengths presented such that both could be seen when a forager was at the decision point. In the orientation platform design (b), foragers had to associate a single colour stimulus to reach the food reward that could be obtained by passing through a hole on one point of the platform. 30 V. Aksoy, Y. Camlitepe / Vision Research 67 (2012) 28–36 connected horizontally to the nest via a hole in the wall of the are- experiments. For intensity discrimination experiments, spectral na at floor level. Each arm of the ‘‘Y’’ extended horizontally for pairs had the same wavelengths but the intensity of unrewarding 200 mm and terminated in a clear perspex feeding box (100 mm one was reduced by 2 log units. Foragers were left free for a week width 150 mm high) from which escape was prevented with a to use both chemical cues and the light stimuli to reach the food. fluon barrier. This arrangement permitted foragers to explore A small quantity of diluted honey and occasionally field collected boxes and return to nest. The foragers had to walk along the floor dead insects were placed in the feeding box to encourage foragers of the Y-maze towards the decision point (where the maze forked), to visit it. For the ensuing week, the Y maze and the orientation which allowed them to see both spectral stimuli at the same time. platform were replaced with a clean one 10 times a day in order The spectral stimuli delivered to the Y-maze were produced by to remove the chemical cues and reinforce the foragers to use direc- light boxes attached to the backside of the feeding boxes. The light tional information of spectral stimuli only. A fresh feeding box was box contained a halogen lamp (Philips Focusline 24 V-250 W, Eind- fitted each day. Training procedure was also resumed between the hoven, The ) and had a built-in ventilator to remove test periods to maintain foraging traffic and reinforcement. the heat produced by the lamps. Interference bandpass filters with 10 nm bandwidth (Thorlabs Inc., Newton, NJ, USA, CWL = 370, 440, 2.5. Tests 540, 590 and 640) were attached to holders in front of the light boxes to obtain monochromatic test stimuli. An adjustable DC Two types of tests were performed; control tests with respec- power supply (Maksimel Inc., model # LPS – 991, Merkez, Ankara, tive training conditions and critical tests, depending on the exper- Turkey) was used to energise the lamps. This power supply with iments, as follows; (i) the stimulus intensities were reduced in the digital panel meters provided precise control of the output intensity response experiments until the distributions of foragers voltage and current with a high stability and very low ripple. Light on the orientation platform was not different from random, (ii) intensity was measured with a calibrated spectroradiometer the intensities of rewarding stimuli were reduced by one log unit (International Light Inc., model # RPS 900, Newburyport, MA, in colour discrimination experiments, and (iii) the 2 log unit train- USA). Absorptive neutral density filters (Thorlabs Inc.) were used ing intensity difference between the stimuli in intensity discrimi- to reduce the intensities of the stimuli by varying factors when nation experiments was decreased to 1 log unit, by increasing needed. Since all tests were performed in darkness, a digital video the intensities of unrewarding ones. camera (Sony DCR-SR90, Tokyo, Japan) with a nightshot vision was Each trial started when a forager entered the orientation plat- used to monitor the ants. form or the maze and lasted when she reached the edge of the plat- The orientation platform consisted of a circular plastic vessel form or entered one of the feeding boxes. Foragers spending more (170 mm diameter) connected to the nest via silicon pipe than 2 min on the platform and inside the maze without any choice (Fig. 1b). The pipe was fixed into a hole at the centre of the plat- were not included in the analysis. Their possible interpretation as form. Escape from the platform was prevented by fluon coating social facilitation was eliminated by the observation that each for- its walls. A short plastic pipe attached in a hole in the vessel wall ager entered and traversed the orientation platform and the maze allowed foragers to access a feeding box. Light boxes were placed alone. The angle at which foragers reached the edge of the orienta- behind the feeding boxes. When a forager exited the silicon pipe tion platform and their tracks during tests were recorded and and reached on the platform she could see the light beam passing transferred in computer media and combined with a program through the filter. (Macromedia Freehand 10.0). Foragers found in either box in the Y maze were recorded. All tested foragers were gently removed 2.3. Elimination of cues by a paint brush and kept in a moist box and put back on their nests when each experimental condition lasted. To be statistically All training and tests were performed in darkness to eliminate robust, at least 30 individuals were used for each test. The distribu- the use of possible visual cues. This also made the foragers posi- tion of compass angles of foragers on the orientation platform was tively phototactic. Since ants can orient themselves using magnetic analysed using a circular statistical method and the distribution of fields (Camlitepe et al., 2005; Camlitepe & Stradling, 1995) and kin- foragers in Y maze was analysed by G-test (Zar, 1984). esthetic cues (Aksoy & Camlitepe, 2005) test stimuli were inter- changed between the arms of the Y-maze after each 15th 3. Results forager. Light boxes and so the stimuli on the orientation platform were also reversed by 90° during tests. Prior to each test, food was 3.1. Intensity response experiments removed and feeding boxes were replaced with clean empty ones. In addition, Y-mazes were replaced with the new ones and the ori- The results of control tests showed that the training intensity (I) entation platform was wiped with alcohol after each tenth forager of all wavelengths caused a significant homeward orientation in order to eliminate any kind of possible chemical cues. The exper- response by foragers (Figs. 2 and 3). The mean vector angle for each imental set-up was placed on a wooden support and levelled to trial was inside the 95% confidence interval. In critical tests, forag- preclude the use of gravitational cues. Therefore, on entering the ers continued to orient to 370 and 540 nm when the intensity maze and reaching on the orientation platform, foragers were de- decrease was 1 log and 2 log units, respectively, but below these nied any point of reference on which to orient. values they dispersed randomly (Figs. 2b and 3b). Foragers’ significant orientation to 440 and 640 nm disappeared when their 2.4. Training intensities were decreased by 1 log unit (Figs. 2d and 3d). Thus, the minimal intensity values that elicited a significant homeward Foragers were trained food rewarded to various spectral stimuli orientation for present experimental conditions were 9 10 for each experimental condition and then tested. Spectral stimuli 1.1 10 photons for 540 nm, 1.1 10 photons for 370 nm and 11 used in intensity threshold and colour vision experiments were 1.1 10 photons for 440 and 640 nm. set to have equal training intensities (I = 1.1 1011 photons). Foragers were trained in turn to 370, 440, 540 and 640 nm for 3.2. Colour vision experiments intensity response experiments and to pairs of 370 vs 540 nm; 440 vs 540 nm; 540 vs 370 nm and 640 vs 540 nm, the former Foragers significantly discriminated 370 nm from 540 nm, and wavelengths being always food rewarding one, for colour vision vice versa, both in control and critical tests (Fig. 4a, b, d, and e). V. Aksoy, Y. Camlitepe / Vision Research 67 (2012) 28–36 31

Fig. 2. Angular distributions and tracks of foragers on the orientation platform during intensity response experiments with 370 and 440 nm. Only the results of control and the last critical tests with which homeward orientations of foragers lost were given (a) 370 nm control test; I = 1.1 1011 photons, P < 0.0005, (b) last critical test, I/ 25 = 4.4 1010 photons, P > 0.05, n.s., (c) 440 nm control test; I = 1.1 1011 photons, P < 0.01, (d) last critical test, I/10 = 1.1 1010 photons, P > 0.05, n.s. The triangle outside each circle indicates the home angle. The dots around the circumference show the actual distribution of angles of foragers. a = mean vector angle; r = mean vector length; u = critical values of the V test; d = deviation values around the 95% confidence interval. The dashed lines denote the 95% confidence interval (c.i.) around each sample mean.

When they were tested to discriminate 440 nm from 540 nm, they conduct the critical test. On the other hand, they significantly failed to do this in the control test (Fig. 4c), a result excluding the discriminated 640 nm from 540 nm in the control test (Fig. 4f) presence of a blue sensitive photoreceptor. We therefore did not but failed in the critical test (Fig. 4g). 32 V. Aksoy, Y. Camlitepe / Vision Research 67 (2012) 28–36

Fig. 3. Angular distributions and tracks of foragers on the orientation platform during intensity response experiments with 540 and 640 nm. Only the results of control and the last critical tests with which homeward orientations of foragers lost were given (a) 540 nm control test; I = 1.1 1011 photons, P < 0.0005, (b) last critical test, I/ 400 = 2.75 108 photons, P > 0.05, n.s., (c) 640 nm control test; I = 1.1 1011 photons, P < 0.0005, (d) last critical test, I/10 = 1.1 1010 photons, P > 0.05, n.s. See legend to Fig. 2 for details.

3.3. Intensity discrimination experiments wavelength but 100 times (2 log unit) dimmer. Since the results of intensity response experiments revealed different intensity val- Foragers were tested to discriminate a stimulus of a constant ues for each wavelength eliciting a significant response, the train- intensity (I) from an unrewarding alternative one of the same ing intensities in this experiment were set to 1.1 1013 photons V. Aksoy, Y. Camlitepe / Vision Research 67 (2012) 28–36 33

Fig. 4. Choice frequencies of foragers in colour discrimination experiments (a) control test; 370 vs 540 nm, G = 6.79, P < 0.01, (b) critical test; 370 vs 540 nm, G = 8.99, P < 0.005, (c) control test; 440 vs 540 nm, G = 1.2, P > 0.05, n.s., (d) control test; 540 vs 370 nm, G = 18.02, P < 0.001, (e) critical test; 540 vs 370 nm, G = 6.79, P < 0.01, (f) control test; 640 vs 540 nm, G = 4.31, P < 0.05, (g) critical test; 640 vs 540 nm, G = 1.05, P > 0.05, n.s. I represents the training intensities and I/10 represents the intensity decrease by 1 log unit.

Fig. 5. Choice frequencies of foragers in intensity discrimination experiments. Test wavelengths and intensities are given under the abscissa. The black bars represent the choice frequencies for the rewarding and the dark and light grey bars represent the choice frequencies for the unrewarding intensities. I = 1.1 1012 photons for rewarding 370 and 540 nm, and I = 1.1 1013 photons for rewarding 440 and 640 nm. I/10 represents intensity decrease by 1 log unit and I/100 represents intensity decrease by 2 log units. (a) G = 0.53, P > 0.05 n.s., (b) G = 14.55, P < 0.001, (c) G = 26.89, P < 0.001, (d) G = 4.93, P < 0.05, (e) G = 3.39, P > 0.05, n.s., (f) G = 14.55, P < 0.001, (g) G = 2.15, P > 0.05, n.s., (h) G = 0.53, P > 0.05 n.s.

for 440 and 640 nm, and 1.1 1012 photons for 370 and 540 nm. In methods (constant threshold response and flickering colourime- the control tests, foragers could discriminate the 2 log unit inten- try), Mazokhin-Porshnyakov (1974) offered a trichromacy for this sity difference for all wavelengths except for 640 nm (Fig. 5). In a species at 375, 470 and 510 nm. However, electrophysiological subsequent test where the intensity difference was decreased to studies yield results on whether an has, in theory, physio- 1 log unit, foragers failed in all wavelength pairs except 440 nm logical mechanisms that could be used in colour vision or not. (Fig 5c). Yet, true colour vision can only be shown by behavioural colour discrimination experiments by which the animal is tested for its 4. Discussion discrimination ability between two lights of different spectral com- position, regardless of their relative intensity (Kelber, Vorobyev, & The results clearly demonstrated that F. cunicularia foragers Osorio, 2003; Menzel, 1979). An investigation of literature on possess a dichromatic colour vision based on UV and green photo- spectral sensitivities of ants yields similar contradictory results receptor inputs (Fig. 4). By contrast, using electrophysiological between behavioural and physiological applications, about the 34 V. Aksoy, Y. Camlitepe / Vision Research 67 (2012) 28–36 number of sensitivity peaks even for the same species. For in- stance, in C. bicolor workers, spectral sensitivity function tests pointed out three peaks at 350, 500–520 and 600 nm whereas the spectral discrimination function tests indicated two peaks at 380 and 550 nm in the same study (Wehner & Toggweiler, 1972) In a later study with C. bicolor, Kretz (1979) reported four peaks at 342, 425, 505 and 570 nm for spectral sensitivity function, and three peaks at 382, 449 and 550 nm for wavelength discrimination function. Furthermore, using intracellular techniques, only UV and green photoreceptors were found in C. bicolor (Labhart, 1986; Mote & Wehner, 1980). Nonetheless, taking together the results of our study and those mentioned above, it seems that UV and green pho- toreceptors are common among ant species. Furthermore, the intensity values obtained with 370 and 540 nm stimuli, the latter being the lowest, that caused a significant orientation response on the orientation platform also support the presence of UV and green photoreceptors. The relatively better green sensitivity can simply be due to a possible high number of green photoreceptors compared to UV photoreceptors in the compound eyes and also may be regarded as an indication of importance of input from Fig. 6. Choice frequencies of foragers between two long wavelengths; 590 nm vs 640 nm. (a) Control test, G = 4.93, P < 0.05, foragers could discriminate between the green channel for foragers. The results also yielded evidence prov- two stimuli presented at equal intensities. (b) Critical test I, the intensity of 590 nm ing that their photoreceptors and the central mechanisms process- was reduced by 1 log unit, G = 3.39, P > 0.05, n.s. (c) Critical test II, the intensity of ing spectral inputs from these photoreceptors provided them a 590 nm was reduced by 2 log units, G = 4.93, P < 0.05, significant for the sensitivity also for wavelengths corresponding to blue and red unrewarding wavelength. I = training intensity, I/10 = intensity decrease by 1 log ranges of the spectrum (Figs. 2 and 3). In addition, recent findings unit, I/100 = intensity decrease by 2 log units. which show that fine colour discrimination in ants takes place in UV and green ranges (Camlitepe & Aksoy, 2010) also support the common existence of UV–green photoreceptors in ants. conditions foragers significantly preferred 590 nm (Fig. 6a). In a One of the striking results we obtained was that, foragers re- second test the intensity of 590 nm was decreased by 1 log unit sponded to achromatic cues to some degree. Although they did and foragers’ preference for the stimuli was not different from ran- not pay attention to chromatic cues provided by 440 and dom but when the decrease by 2 log units in a third test, their pref- 640 nm, they responded significantly to these wavelengths on erence was for 640 nm (Fig. 6a and b). We also tested foragers the orientation platform when provided above a critical intensity distributions on the orientation platform in the presence of 590 value. Furthermore, they were successful in the intensity discrim- and 640 nm and found that the mean distributions were statisti- ination task at 440 nm even with a 1 log unit intensity difference cally significant showing that a homeward orientation was estab- (Fig. 5c and d). This clearly indicates that both 440 and 640 nm ex- lished (Fig. 7). It is clear from all these results that responses of cited the photoreceptors, and they were used by the ants to reach foragers to long wavelengths, as in M. stellatorum, is colour blind the food. Intensity discrimination between two stimuli of the same and mediated by achromatic cue perception. As the proposed UV wavelength was previously reported for the desert ant C. bicolor sensitive photoreceptors possibly do not contribute to this long (Kretz, 1979). can pay more attention to chromatic cues wavelength perception, we conclude that foragers responded to when they are presented simultaneously with achromatic ones long wavelengths with a photon catch mechanism by their green and can also respond achromatic cues when they are presented receptors and oriented accordingly. alone (Kelber, 2005). Thus, the finest intensity discrimination suc- There exist contradictory but not satisfactory data considering cess of foragers at 440 nm among other wavelengths could be ex- responses of ants to long wavelengths. The action spectrum of plained by the possibility that foragers paid more attention to Marak and Wolken (1965) obtained with S. saevissima workers re- achromatic cues when they were required to discriminate between vealed a peak at 620 nm which was related with possible effects of two 440 nm stimuli providing them no chromatic perception. screening pigments and ommochroms. The spectral sensitivity It has been long known that ants, as in many insects, are insen- functions obtained for C. bicolor by Wehner and Toggweiler sitive to long wavelengths within the red (in terms of human vi- (1972) and Kretz (1979) revealed peaks at 600 and 570 nm respec- sion) part of the spectrum. However, recent findings of long tively, but these wavelengths were not discriminated from alterna- wavelength sensitivity and discrimination for bees in longer wave- tives in a colour vision context. A similar long wavelength lengths bring out the need of questioning red-blindness in insects sensitivity has recently been shown for Formica pratensis (Camli- again (Chittka & Waser, 1997; Martínez-Harms et al., 2010; Reisen- tepe et al., 2006) and Cataglyphis aenescens workers (Aksoy & Cam- man & Giurfa, 2008). All bees tested so far for their spectral sensi- litepe, unpublished data). It is thus apparent that benefiting red tivities have been shown to have sensitivity for longer wavelengths light for observation simplicity when testing insects in laboratory (Peitsch et al., 1992). For instance, although honeybee L-receptor conditions should be carefully evaluated both in experiments and (green receptor) maxima peak is at 550 nm their sensitivity at long in interpreting the results. wavelength part of the spectrum extends to 650 nm (Chittka & The study of Möller (2002) involves important suggestions Waser, 1997). about landmark navigation of insects which might account for Considering the significant responses of foragers to 640 nm on the prevalence of UV–green dichromacy among ants. A contrast the orientation platform, we designed a further test in which we mechanism involving UV and green receptors of insect eyes could trained and tested foragers’ responses in a discrimination task be- guarantee a robust separation between natural objects as fore- tween a rewarding 590 nm and an unrewarding 640 nm of the ground and the sky as background. UV and green receptors also same intensities (I). A similar experiment was performed on the contribute some other visual tasks in insects. UV receptors were diurnal moth species Macroglossum stellatorum by Kelber and shown to play role in polarised light sensitivity of honeybees Henique (1999). In the control test performed with training (Rossel & Wehner, 1984) and the workers of the desert ant V. Aksoy, Y. Camlitepe / Vision Research 67 (2012) 28–36 35

Fig. 7. Angular distributions and tracks of foragers on the orientation platform with 590 and 640 nm. The training and test intensities are same (I = 1.1 1011 photons) for both stimuli. (a) 590 nm; P < 0.005, (b) 640 nm; P < 0.005. A homeward orientation was established for both 590 and 640 nm.

C. bicolor (Duelli & Wehner, 1973; Labhart, 2000) and green recep- 2000). Colour vision can be vital for ants during their outbound tors to play a role in motion related tasks and target detection in trips in their environments where they also face different objects honeybees (Giurfa et al., 1997; Lehrer, 1994). Perception of achro- with different spectral compositions. On the other hand, compared matic cues in honeybees is also mediated by green receptors (Leh- with trichromatic vision, dichromatic vision entails a loss of hue rer, 1994). The fact that all ommatidia in honeybee compound eyes discrimination and results in a reduced colour gamut. However, have six green receptors underlines the importance of this input Camlitepe and Aksoy (2010) have recently shown that F. cunicular- channel in different visual performances (Wakakuwa et al., ia and C. aenescens workers can discriminate similar wavelengths 2005). Green receptors are also in much number compared to UV in UV and green ranges. Therefore, F. cunicularia foragers must have (6:2) in compound eyes of F. polyctena workers (Menzel, 1973). a neural representation of colour at higher levels to account for Although the number of different photoreceptor types in a tri- their spectral sensitivity and learning abilities. We believe that fu- chromatic insect, such as honeybees, provide that insect a more ture electrophysiological and behavioural studies on other ant spe- successful colour vision compared to a dichromatic one, environ- cies will reveal considerable data to demonstrate the prevalence mental conditions affect the potential advantages or disadvantages and diversity of ant colour vision among Hymenopterans. of one type of colour vision (tri- or dichromatic) over the other. In the natural world, the amount of ambient light is prone to changes Acknowledgments by effects caused by weather, time of day, location in the canopy and amount of surrounding foliage (Endler, 1993). In decreasing This research was supported by the Research Fund of Trakya ambient light conditions, achromatic vision becomes relatively University (TUBAP Project # 703). more important compared to chromatic vision, which is evident in the nocturnal adaptations of many animals (Melin et al., References 2007). Therefore, in such light conditions dichromatic vision may Aksoy, V., & Camlitepe, Y. (2005). Use of idiothetic information for left/right turning be especially advantageous over trichromatic vision for achromatic memory by the ant Formica pratensis. Biologia, 60(2), 197–200. tasks. In a recent study, Caine, Osorio, and Mundy (2010) argued Briscoe, A., & Chittka, L. (2001). The evolution of color vision in insects. Annual that dichromacy in marmosets is an advantage over trichromacy Review of Entomology, 46, 471–510. when foraging in low light intensity conditions since achromatic Caine, N. G., Osorio, D., & Mundy, N. I. (2010). A foraging advantage for dichromatic marmosets (Callithrix geoffroyi) at low light intensity. Biology Letters, 6, 36–38. cues are paid more attention by dichromats than trichromats. An- Camlitepe, Y., & Aksoy, V. (2010). First evidence of fine colour discrimination ability other advantage attributed to dichromats is that they perform bet- in ants (Hymenoptera, Formicidae). The Journal of Experimental Biology, 213, ter in detecting colour-camouflaged objects against background 72–77. Camlitepe, Y., Aksoy, V., Uren, N., Yilmaz, A., & Becenen, I. (2005). An experimental (Morgen, Adam, & Mollon, 1992). It has been hypothesised that analysis on the magnetic field sensitivity of the black-meadow ant Formica colour vision systems are a result of evolutionary adaptations for pratensis Retzius (Hymenoptera: Formicidae). Acta Biologica Hungarica, 56(3–4), object discrimination (Lythgoe & Partridge, 1989). Most mammals 215–224. Camlitepe, Y., Aksoy, V., Uren, N., Türkog˘lu, K. A., & Yilmaz, A. (2006). Siyah sırtlı are dichromats and live in forest habitats where they have to see orman karıncası Formica pratensis (Hymenoptera: Formicidae)’de kırmızı dalga and discriminate leaves, soil, barks and stones (Chiao et al., boyu duyarlılıg˘ı, XVIII. Ulusal Biyoloji Kongresi, Kusßadası – AYDIN. 36 V. Aksoy, Y. Camlitepe / Vision Research 67 (2012) 28–36

Camlitepe, Y., & Stradling, D. J. (1995). Wood ants orient to magnetic fields. Mazokhin-Porshnyakov, G. A., & Trenn, V. (1972). Electrophysiological study of Proceedings of the Royal Society of London Series B – Biological Sciences, 261, vision in ants. Zoologichesky Zhurnal, 51(7), 1007–1017. 37–41. Melin, A. D., Fedigan, L. M., Hiramatsu, C., Sendall, C. L., & Kawamura, S. (2007). Chiao, C.-C., Vorobyev, M., Cronin, T. W., & Osorio, D. (2000). Spectral tuning of Effects of colour vision phenotype on insect capture by a free-ranging dichromats to natural scenes. Vision Research, 40, 3257–3271. population of white-faced capuchins, Cebus capucinus. Animal Behaviour, Chittka, L., Beier, W., Hertel, H., Steinmann, E., & Menzel, R. (1992). Opponent colour 73(1), 205–214. coding is a universal strategy to evaluate the photoreceptor inputs in Menzel, R. (1973). Evidence for color receptors in the Hymenoptera eye obtained hymentoptera. Journal of Comparative Physiology A, 170, 545–563. from selective adaptation experiments. T-I-T Journal of Life Sciences, 3, 95–100. Chittka, L., & Waser, N. M. (1997). Why red flowers are not invisible to bees. Israel Menzel, R. (1979). Spectral sensitivity and colour vision in invertebrates. In H. Journal of Plant Science, 45, 169–183. Autrum (Ed.), Handbook of sensory physiology (Vol. 6, pp. 503–580). Berlin, Duelli, P., & Wehner, R. (1973). The spectral sensitivity of polarized light orientation Heidelberg, New York: Springer. in Cataglyphis bicolor (Formicidae, Hymenoptera). Journal of Comparative Menzel, R., & Backhaus, W. (1991). Colour vision in insects. In P. Gouras (Ed.), Vision Physiology, 86, 37–53. and visual dysfunction: The perception of colour (pp. 262–288). London: Mac- Endler, J. (1993). The colour of light in forests and its implications. Ecological Millan. Monographs, 63, 1–27. Menzel, R., & Blakers, M. (1975). Functional organization of an insect ommatidium Giurfa, M., Vorobyev, M., Brandt, R., Posner, B., & Menzel, R. (1997). Discrimination with a fused rhabdom. Cytobiology, 11, 279–298. of coloured stimuli by honeybees: Alternative use of achromatic and chromatic Menzel, R., & Knaut, R. (1973). Pigment movement during light and chromatic signals. Journal of Comparative Physiology A, 180, 235–243. adaptation in the retinula cells of (Hymenoptera, Formicidae). Kelber, A. (2005). Alternative use of chromatic and achromatic cues in a hawkmoth. Journal of Comparative Physiology, 86, 125–138. Proceedings of the Royal Society of London Series B – Biological Sciences, 273, Möller, R. (2002). Insects could exploit UV–green contrast for landmark navigation. 2143–2147. Journal of Theoretical Biology, 214(4), 619–631. Kelber, A., & Henique, U. (1999). Trichromatıc colour vision in the hummıngbird Morgen, M. J., Adam, A., & Mollon, J. D. (1992). Dichromats detect colour- hawkmoth, Macroglossum stellatarum L. Journal of Comparative Physiology, 184, camouflaged objects that are not detected by trichromats. Proceedings of the 535–541. Royal Society of London Series B – Biological Sciences, 248, 291–295. Kelber, A., Vorobyev, M., & Osorio, D. (2003). Animal colour vision – Behavioural Mote, M. I., & Wehner, R. (1980). Functional characteristics of photoreceptors in the tests and physiological concepts. Biological Reviews, 78, 81–118. compound eye and ocellus of the desert ant, Cataglyphis bicolor. Journal of Kiepenheuer, J. (1968). Farbunterscheidungsvermgen bei der roten Waldameise, Comparative Physiology, 137, 63–71. Formica polyctena. Zeitschrift fur Vergleichende Physiologie, 57, 409–411. Peitsch, D., Feitz, A., Hertel, H., de Souza, J., Ventura, D. F., & Menzel, R. (1992). The Kretz, R. (1979). A behavioural analysis of colour vision in the ant Cataglyphis bicolor spectral input systems of hymenopteran insects and their receptor-based (Hymenoptera, Formicidae). Journal of Comparative Physiology, 1311, 217–233. colour vision. Journal of Comparative Physiology A, 170, 23–40. Kühn, A. (1924). Versuche über das Unterscheidungsvermögen der Bienen und Reisenman, C. E., & Giurfa, M. (2008). Chromatic and achromatic stimulus Fische für Spektrallichter. Nachrichten der Gesellschaft für Wissenschaften, 1, 66– discrimination of long wavelength (red) visual stimuli by the honeybee Apis 71. mellifera. –Plant Interactions, 2, 137–146. Labhart, T. (1974). Behavioural analysis of light intensity discrimination and Rossel, S., & Wehner, R. (1984). Celestial orientation in bees: the use of spectral cues. spectral sensitivity in the honeybee, Apis mellifera. Journal of Comparative J. comp. Physiol. A, 155, 605–613. Physiology, 95, 203–216. Roth, H., & Menzel, R. (1972). ERG of Formica polyctena and selective adaptation. In Labhart, T. (1986). The electrophysiology of photoreceptors in the different eye R. Wehner (Ed.), Inf. proc. vis. syst. arthr. (pp. 177–181). Berlin, Heidelberg, New regions of the desert ant, Cataglyphis bicolor. Journal of Comparative Physiology A, York: Springer. 158, 1–7. Skorupski, P., & Chittka, L. (2010). Differences in photoreceptor processing speed for Labhart, T. (2000). Polarization-sensitive interneurons in the optic lobe of the desert chromatic and achromatic vision in the bumblebee, Bombus terrestris. Journal of ant Cataglyphis bicolor. Naturwissenschaften, 87, 133–136. Neuroscience, 30, 3896–3903. Lehrer, M. (1994). Spatial vision in the honeybee: The use of different cues in Tsuneki, K. (1953). On colour vision in two species of ants, with special emphasis on different tasks. Vision Research, 34, 2363–2385. their relative sensitivity to various monochromatic lights. Japanese Journal of Lubbock, J. (1929). Ants, bees, and wasps. London, England: Kegan. Zoology, 11, 187–221. Lythgoe, J. N., & Partridge, J. C. (1989). Visual pigments and the acquisition of visual Turner, C. H. (1907). The homing of ants: An experimental study of ant behaviour. information. Journal of Experimental Biology, 146, 1–20. Journal of Comparative Neurology and Psychology, 17, 367–434. Marak, G. E., & Wolken, J. J. (1965). An action spectrum for the fire ant Solenopsis von Frisch, K. (1914). Der Farbensinn und Formensin der Biene. Zoologische saevissima. Nature, 205, 1328–1329. Jahrbucher-Abteilung fur Allgemeine Zoologie und Physiologie der Tiere, 35, 1–188. Martínez-Harms, J., Palacios, A. G., Márquez, N., Estay, P., Arroyo, M. T. K., & Wakakuwa, M., Kurasawa, M., Giurfa, M., & Arikawa, K. (2005). Spectral Mpodozis, J. (2010). Can red flowers be conspicuous to bees? Bombus dahlbomii heterogeneity of honeybee ommatidia. Naturwissenschaften, 92, 464–467. and South American temperate forest flowers as a case in point. The Journal of Wehner, R., & Toggweiler, F. (1972). Verhaltensphysiologischer Nachweis des Experimental Biology, 213, 564–571. Farbensehens bei Cataglyphis bicolor (Formicidae, Hymenoptera). Journal of Mazokhin-Porshnyakov, G. A. (1974). Investigations on the vision of ants. In G. A. Comparative Physiology, 77, 239–255. Horridge (Ed.), The compound eye and vision of insects (pp. 114–120). Oxford: Zar, J. H. (1984). Biostatistical analysis (2nd ed.). Englewood Cliffs, NJ: Prentice-Hall. Clarendon Press.