Studies on Light Sensitivity in Butterflies

Study on insect color vision is pioneered by Karl von Frisch. Frisch demonstrated that honeybees have color vision when searching for food: honeybees can even see UV as a separate color, but instead they don’t discriminate red from gray. Since the impact of Frisch’s work was so strong that similar study in other insects had become less intense. Probably therefore, it had mistakenly become more or less a general understanding that all insects are not able to discriminate red. However, many butterflies collect nectar from red flowers, which honeybees seldom visit. Do butterflies see colors, including red? First convincing demonstration of red sensitive photoreceptors in the butterfly compound eye appeared in 1983 (Matic, J Comp Physiol A 152:169-182, 1983), which directly and strongly stimulated us to initiate the study of the compound eye of the Japanese yellow swallowtail butterfly, Papilio xuthus . By combining electrophysiological, histological, and molecular biological methods, we analyzed cellular organization of the Papilio retina (Arikawa, 2003). We so far identified six classes of spectral receptors (UV, violet, blue, green, red, and broadband receptors) there (Arikawa et al. , 1987; Arikawa et al. , 2003). We then localized these receptors in the ommatidia, the building blocks of the compound eye, each containing nine photoreceptor cells. Quite unexpectedly, we found that the ommatidia are divided into three types in terms of the spectral receptor classes they contain. The three spectrally heterogeneous ommatidia distribute rather randomly over the compound eye (Arikawa and Stavenga, 1997). In the course of the physiological and anatomical studies of the Papilio retina, we found several interesting phenomena. For example, the broadband receptors coexpress green absorbing and red absorbing visual pigments both contribute to generate receptor potential (Arikawa et al. , 2003; Kitamoto et al. , 1998). This indicates that a dogma in vision physiology, which is that a photoreceptor cell express a single type of visual pigment, needs adjustment. We also found a reverse case: violet receptors express UV absorbing visual pigment that is functional also in UV receptors (Kitamoto et al. , 2000). What constructs the violet spectral sensitivity from UV visual pigment is 3-hydroxyretinol that is somehow accumulated in the distal portion of the ommatidia. 3-hydroxyretinol there absorbs UV and thus acts as an UV absorbing filter (Arikawa et al. , 1999). An extreme case was found in the eye of the small white butterfly, Pieris rapae , where green, red, and deep-red receptors all share a green-absorbing visual pigment. Red and deep-red pigments around the rhabdom act as spectral filters that produce distinct spectral sensitivities from a single visual pigment (Wakakuwa et al. , 2004). Do butterflies really see colors? To answer the question we established a protocol of behavioral experiments. We trained butterflies to take sucrose solution on a paper patch of certain color in a small cage. We let the trained butterflies select the training color among different colors or different shades of grays, and demonstrated that foraging Papilio have real color vision when searching for food (Kinoshita et al. , 1999), and have even color constancy (Kinoshita and Arikawa, 2000). Another behavioral experiment where we use monochromatic stimuli has suggested that the Papilio color vision system is tetrachromatic based on the UV, blue, green, and red receptors.