(Hypsipetes leucocephalus nigerrimus, Pycnonotidae)
24
Abstract
Sexual dichromatism is an important proxy for the intensity of sexual selection, yet related studies in birds based on museum specimens or conspicuous visual traits in live animals may have led to an
underestimation of the intensity and complexity of sexual selection.
Using the Himalayan black bulbul (Hypsipetes leucocephalus nigerrimus) which is sexually monomorphic to the human eye, I investigated the extent of overall bodily sexual dichromatism. I measured the reflectance- both within human visual perceptive range and in the ultra-violet range- of two carotenoid-based parts and eight dull, melanin-based parts for each individual live bird or museum skin sampled. I found that males had redder beaks, brighter tarsi and darker plumage than females. These are perceptible to the bird according to a model of color discrimination thresholds, suggesting the existence of multiple cryptic sexually selected traits within the species. I also found significant degradation of the color in skin specimens compared with that in live birds, indicating that sexual dichromatism could be underestimated by using skin specimens alone.
Keywords: carotenoid-based characteristics, cryptic sexual dichromatism, melanin-based plumage, Himalayan black bulbul (Hypsipetes
leucocephalus nigerrimus), sexual dichromatism
25
Introduction
One of the most robust and widely used indexes of the intensity of sexual selection in birds is sexual dichromatism, in which the male is typically brighter or more colorful than the female or has more distinguishing features (e.g., Owens and Hartley 1998, Seddon et al. 2013). Although intersexual differentiation in mating behavior, habitat preference or predator avoidance can also promote the evolution of coloration, sexual dichromatism is considered to be driven mainly by female preference or male-male competition (Andersson 1994) and to function in sexual recognition, individual quality assessment and sexual attraction (Dale 2006). Still, several pitfalls in studies of sexual dichromatism may have led to an overall underestimation of sexual selection in birds. Most significantly, these studies have mainly focused on conspicuous differences (e.g., Bortolotti et al. 1996, Eaton 2005, Gray 1996) in the range 400 to 700nm perceptible to human vision (Cuthill et al. 1999, Neitz and Jacobs 1986). However, birds have a wider visual sensory range (300 to 700nm), and can detect intersexual differences in ultra-violet (UV; 300 to 400nm, Chen et al. 1984). Therefore, UV coloration could be also used as the signal (e.g., Alonso-Alvarez et al.
2004, Siefferman and Hill 2005) or target (Bennett et al. 1997) for mate choice. With the aid of spectrometers, several avian species presumed monochromatic have been found to have dichromatic UV coloration (e.g., Igic et al. 2010, Mays Jr et al. 2004), but more studies are needed to evaluate the prevalence of UV-dichromatism.
Also underrepresented in avian sexual dichromatism studies is the examination of melanin-based coloration, which appears dull to humans
26
but may still carry signals of individual quality to birds given their superior vision. Melanin-based characteristics have been found to be associated with individual qualities such as social rank, aggressive behavior and immunocompetence and should be no less important targets for sexual selection than carotenoid-based characteristics (reviewed in McGraw 2006b, Kingma et al. 2008, Tarof et al. 2005). Moreover, melanin deposition appears to be controlled by genes and not easily affected by environmental factors such as diet (Fox 1976, Buckley 1989), whereas the expression of carotenoids depends on nutritional status and foraging ability (Hill 1992, Nolan et al. 1998, Thompson et al. 1997). The sexual selection pressures on these traits might differ from those on other types of trait existing in the same organism. However, little attention has been paid to the relative contribution of the two pigment-based
colorations within the same species.
The use of museum skin specimens in studies of avian coloration could also lead to the underestimation of sexual dichromatism. The concern that specimens’ feather color might fade over time has been noted in several studies: the color degradation might be species dependent and also determined by when the specimen was collected (Doucet and Hill 2009, McNett et al. 2005, Pohland and Mullen 2006). It has been shown that color fading is significant for museum skin specimens collected more than 50 years previously (Pohland and Mullen 2006). But the level of color degradation of newly collected museum specimens has been controversial (Doucet and Hill 2009, McNett et al. 2005, Pohland and Mullen 2006).
In this study, I used a spectrometer to study sexual dichromatism in a
27
passerine, the Himalayan black bulbul (H. leucocephalus nigerrimus), which is sexually monomorphic to the human eye: both sexes are entirely covered by black plumage with a grey patch on their wings (both
melanin-based) and a red beak and tarsus (both carotenoid-based
(McGraw 2006a). I tested whether intersexual plumage-color differences would be perceptible to the bulbul itself with the Vorobyev-Osorio color discrimination model which is based on the avian tetrahedral color space (Eaton 2005, Stoddard and Prum 2008). I show that sexual dichromatism does exist in the Himalayan black bulbul, providing insights into the potential functional roles of melanin-based and carotenoid-based characteristics in the species. I also found significant fading of museum skin specimens less than five years old. This raises concerns about the use of recently collected skin specimens for study of avian cryptic sexual dichromatism.
Materials and methods Study Species
The Himalayan black bulbul (H. leucocephalus nigerrimus) is a widely distributed species inhabiting broadleaf evergreen and mixed deciduous forests, groves, clearings and edges. A total of 112 live individuals were bought from the pet-shop (San Xing Bird Shop, Taipei,
25.034398,121.504444) during the non-breeding seasons of 2008, 2009 and 2011. The birds were all captured from the southern mountain areas in Taiwan. A blood sample was taken from each bulbul for molecular sex typing before proceeding to color quantification. I also examined 37 specimens from the archives of Taiwan’s National Museum of Natural
28
Science (female: 5, male: 11) and Endemic Species Research Institute (female: 7, male: 14), all of which had been collected within the previous 15 years.
Molecular Sex Typing
Gross DNA was extracted from blood samples with traditional proteinase K digestion followed by LiCl extraction (modified from the procedure of Gemmell and Akiyama 1996). Extracted DNA was resuspended in ddH2O and stored at −20°C. Less than 100 ng of genomic DNA was added to 12.5 μL of PCR (polymerase chain reaction) mix containing 0.5 mM of each of the dNTPs, 0.3 μM of each PCR primer (2550F / 2718R,
Fridolfsson and Ellegren 1999), 10 mM Tric-HCL, 50 mM KCL, 1.5 mM of MgCl2 and 0.4 U of Taq DNA polymerase (Amersham Biosciences).
The PCR profile was 94°C for 3 min, followed by 40 cycles of 95°C for 20 s, 46°C for 30 s and 72°C for 40 s, finished at 72°C for 2 min. The PCR reactions were carried out in iCyclers (Bio-Rad, Hercules, CA, USA). After PCR reactions, I conducted electrophoresis with 1.2%
agarose gel to determine the sex. In total, 55 male and 57 female live bulbuls were identified.
Color Measurement
For each individual, the reflectance of ten body regions including two carotenoid-based parts- the beak and tarsus- and eight melanin-based parts- the forehead, nape, back, breast, belly, tail, remige and scapular feathers- were measured by an USB2000 spectrometer (Ocean Optics) with a HL2000 deuterium-halogen light source (Ocean Optics). A
29
R600-7-UV/125F probe (Ocean Optics) was held perpendicular to the surface of the feathers with a cylindrical cap at the end to standardize measuring distance (5 mm) and to shield ambient light. To calculate relative reflectance, a white standard (Labsphere) was used. To collect the dark reference, the light source was capped by a black plastic plate. Each part was measured three times to calculate repeatability (repeatability >
90% (Lessells and Boag 1987). Due to the obvious fading of
carotenoid-based coloration, I did not score the coloration of beak and tarsus in skin specimens. I measured the coloration after checking that there was no obvious stain or abrasion on the surface in order to reduce errors of diminished light reflectance.
Color Quantification
I used a combination of colorimetric variables to quantify coloration.
These included hue, total brightness and chroma (Montgomerie 2006).
Hue was calculated for beak and tarsus by finding the wavelength of the mean of maximum and minimum reflectance values in the wavelength range 550 to 700nm. Total brightness was calculated for all parts by averaging the reflectance from 300 to 700 nm. Two kinds of chroma were calculated. One was chromaRED, calculated as the proportion of
reflectance from 550 to 700 nm in the total brightness for beak and tarsus.
The other was chromaUV, calculated as the proportion of reflectance from 300 to 400 nm on the total brightness for all parts.
Color Discrimination
To distinguish between the hue, brightness and chroma of the two sexes, I
30
used two-way ANOVA to compare the male and female average
measurements by considering the cofactor of different examining year. I also used two-way ANOVA to test whether the date of sample collection (in years) had any significant effect on these colorimetric variables. I also compared the same colorimetric variables between skin specimens and the live birds to examine the extent of color by using two-way ANOVA.
Because hue, brightness and chroma were compared for each of two carotenoid-based and eight melanin-based body parts, I applied a Bonferroni adjustment for multiple comparisons which reduced the p value from 0.05 to 0.025 in red parts and to 0.00625 in black plumage.
I calculated the variability (standard deviation) of brightness within each sex to test whether the divergence of color differences within the females was different from that within the males at different parts. I also used two-way ANOVA to test whether sexually dichromatic parts had higher variability than non-sexually dichromatic parts.
In addition, considering the different spectral sensitivity of the four avian cone types, I mapped the spectra onto Goldsmith’s tetrahedral color space system (Goldsmith 1990) that has recently been recommended for analyzing avian coloration (Eaton 2005, Stoddard and Prum 2008, Stoddard and Prum 2011). I converted the spectrum measured into points within a tetrahedron in which the vertices correspond to exclusive
stimulation of the ultraviolet (UV)-, blue (B)-, green (G)- and red
(R)-sensitive cones in the avian eye. The quantum catch of each receptor is as follows:
Qi = ∫λRi(λ)S(λ)I(λ)dλ,
where λ denotes wavelength, Ri(λ) is the spectral sensitivity of cone cell
31
type i (i from 1 to 4 represent the four cone cells, UVS or VS, SWS, MWS and LWS respectively), S(λ) is the reflectance spectrum of a given feather patch, I (λ) is the irradiance spectrum entering the eye and
integration is over the entire avian visual range(300-700 nm). The program Tetracolorspace (Stoddard and Prum 2008) was used for spectrum conversion, and I chose the average spectral sensitivity curves of UVS-type retinas (Endler and Mielke JR 2005) as the candidate avian vision in this study. After calculating the Qi, I calculated discriminability of color for each pair of average males and females in different body patches using the Vorobyev-Osorio color discrimination model (Vorobyev and Osorio 1998, Vorobyev et al. 1998). The model calculates a distance in avian color space (ΔS) defined by the quantum catch of each receptor type (i.e., cone cell) in the avian retina (Eaton 2005). To calculate ΔS, I used the following formula:
(ΔS)2 = [(ω1ω2)2(Δf4-Δf3)2 + (ω1ω3)2(Δf4-Δf2)2 + (ω1ω4)2(Δf3-Δf2)2 + (ω2ω3)2(Δf4-Δf1)2 + (ω2ω4)2(Δf3-Δf1)2 + (ω3ω4) 2 (Δf2-Δf1)2] / [(ω1ω2ω3)2 + (ω1ω2ω4)2 + (ω1ω3ω4)2 + (ω2ω3ω4)2]
where ωi is the constant noise-to-signal ratio (Weber fraction) for receptor type i, which is based here on empirical estimates from the Pekin robin (Leiothrix lutea, ω4=0.05, following the ratio of the numbers of cones (UV: S: M: L= 1:2:2:4). fi is proportional to the natural logarithm of the respective receptor quantum catches, which are normalized against an adapting background (equation 2 and 3 of (Vorobyev et al. 1998)). Δfi is the difference between the signals in receptor i between the stimuli (two colors). When ΔS is below a threshold value 1, colors are assumed to be indistinguishable.
32
Results
The average spectrums of two sexes were similar in shape but different in total reflectance (Fig. 2.1). In the carotenoid-based beak and tarsus, the spectrums show two peaks at wavelengths ranging from 300 to 400 nm and 600 to 700 nm, which are the reflectance ranges of UV light and carotenoid feathers respectively. Conversely, the spectrums for the melanin-based parts were almost flat but with a slight rise in the UV section. Among the colorimetric variables, the hue and the total brightness were different between sexes in two places. The
carotenoid-based beaks of males had higher hues than those of females (male 590.25±0.62 nm, female 587.95±0.71nm; least square mean ± SE;
Fig. 2.2., two-way ANOVA, p = 0.016). At the melanin-based belly, males had lower total brightness than females (male 4.30±0.21%, female:
5.16±0.20%, p=0.004, Fig.2.3). As for skin specimens, all parts were the same in the two sexes (S 2.1 Table).
Applying the Vorobyev-Osorio color discrimination model, more parts were found to be significantly dichromatic in live male and female birds, including the carotenoid-based beak and tarsus and the
melanin-based belly, remige and tail (Table 2.1). In museum skin
specimens, the belly and scapular- in addition to breast- were also found to be sexually dichromatic. As such, different subsets of body parts were found to be sexually dichromatic in live birds and museum skin
specimens (Table 2.1).
Color comparisons between live birds and skin specimens showed
33
coloration fading in several parts. Live birds had higher brightness in breast and scapular but lower brightness in the tail (Table2.2 a, S2.2 Table); they also had higher chromaUV in every part (Table2.2 b, S2.2 Table b). Although the sampled skin specimens were all collected less than 20 years previously, an effect of specimens’ preserved years was found in the scapular: older specimens showed significantly lower
brightness than more recent ones (specimens 15-10 years old 4.52±0.99%, specimens 10-5 years old 5.13±0.53%, specimens less than 5 years old 7.21±0.69%; F =3.70, p=0.042, S2.2 Table), although the difference is not statistically significant after a Bonferroni correction to account for
multiple comparisons.
Color variabilities of brightness within females were the same with those within males in all parts (Table 2.3). Additionally, the variabilities of brightness of sexually dichromatic parts (carotenoid-based beak, tarsus and melanin-based belly, remige and tail) were larger than those of sexually monochromatic ones (Table 2.3, variances of sexual dichromatic traits 3.67±0.62, variances of non-sexual dichromatic traits 1.06±0.50, Two-way ANOVA with cofactor sex, F=10.663, p=0.005).
Discussion
I have shown significant sexual dichromatism in both carotenoid-based and melanin-based body regions of the Himalayan black bulbul, including in reflectance and spectral shape. Males’ redder beak, brighter tarsus and darker plumage were significantly different enough for birds to
distinguish between them and females, which could provide an insight
34
into this species’ mating behavior. I also show that color degradation could lead to different results on sexually dichromatism in skin specimens and live birds.
The Himalayan black bulbul and most pycnonotids (part of
Pycnonotidae), are dull to humans and listed as monomorphic (Fishpool and Tobias 2005), but my results suggest that the extent of their sexual dichromatism could be underestimated; it is significant but not very large (Table 1; ΔS of Black bird (Turdus merula): 5.56-9.21; ΔS of Black cap (Sylvia atricapilla): 1.48-16.9; ΔS of Greenfinch (Carduelis chloris):
2.26-8.10 (Delhey and Peters 2008), which may be indicative of mild sexual selection. Like most pycnonotids, the Himalayan black bulbul is socially monogamous and provides biparental care (Fishpool and Tobias 2005), personal observation). Dunn et. al. (2010) analyzed more than 1000 species of birds and found lower sexual dimorphism in species with monogamous than with polygynous or lekking mating systems where variance in male mating success is thought to be lower. Nevertheless, other aspects of Himalayan black bulbul and related species’ reproductive biology may contribute to sexual dichromatism; these include the genetic mating system and the parental investments of each sex, which need to be investigated further.
The Himalayan black bulbul’s sexually dichromatic characteristics could be function-signaling and therefore the objects of sexual selection.
In a study of six avian species, Delhey and Peters (2008) found that most function-signaling patches were sexually dichromatic. Sexually
dichromatic traits have been proved to function in quality signaling (e.g., Hill 1996, Walker et al. 2013) and agonistic interactions in several avian
35
species (e.g., Alonso-Alvarez et al. 2004, Bright and Waas 2002, Préault et al. 2002), and are often the object of female choice. Moreover,
brightness varied more in the sexually dichromatic parts of the Himalayan black bulbul (beak, tarsus, belly, remige and tail) than in sexually
monochromatic ones, consistent with theoretical and empirical expectations (Andersson 1994, Delhey and Peters 2008).
Where males are subject to female mate choice, their sexually selected traits are usually more variable than females’ (Andersson 1994, Darwin 1872). The similar variability that I found in both female and black bulbuls’ sexually dichromatic traits suggests that mate choice might be mutual in this species. Whereas studies of sexual selection have mostly focused on female choice and male–male competition, data increasingly shows that males can be choosy and benefit from mating females whose reproductive potential is high (reviewed in Edward and Chapman 2011, e.g., Amundsen and Forsgren 2001, Griggio et al. 2005, Jones et al. 2001).
Kokko and Johnstone (2002) suggested that high species-specific and high sex-specific mate-encounter rates, high cost of breeding (parental investment), low cost of mate searching and highly variable quality of the opposite sex could promote the evolution of choosiness and that the primary determinant of sex roles in mate choice is parental investment.
According to this hypothesis, the sex for which the cost of breeding (mortality during signaling and caring) is the larger should evolve to be choosy. The reproductive biology of Himalayan black bulbuls is unclear, but research on pycnonotids suggests comparable parental care loads between the sexes, and the breeding success is generally low (8.3-15%, Balakrishnan 2010, Fishpool and Tobias 2005) while the rate of predation
36
is high. As such, high cost of breeding and comparable parental care load between the sexes might promote mutual selection in pycnonotids.
My data also suggest the involvement of multiple Himalayan black bulbul ornaments in sexual selection - as both carotenoid-based and melanin-based characteristics were found to be sexually dichromatic.
Studies have shown females choosing mates based on multiple sexual ornaments (Chaine and Lyon 2008, Doucet and Montgomerie 2003);
multiple ornaments provide females with different kinds of information in different stages of mate choice (Borgia 1995), or function as redundant signals to improve the accuracy of mate assessment (Johnstone 1994, Moller and Pomiankowski 1993).
Different sets of sexually dichromatic parts were found in live birds and museum skin specimens, and significant degradation of color- whether pigment-based or structural- was found in skin specimens, some of which had been preserved for less than 5 years. These results suggest that the use of skins in avian coloration study may be error-prone, contradicting the previous finding that melanin- and carotenoid-based skins colors remain the same for at least 50 years after preservation (Armenta et al. 2008). Conversely, my results corroborate the conclusion drawn in a study comparing live and skin long-tailed manakins that significant differences in colorimetric variables were attributable to the age of specimens (Doucet and Hill 2009). They also agree with another study that found UV color degradation in preserved skin specimens of some 300 bird species throughout Europe and the USA (Pohland and Mullen 2006). There are many possible reasons for color degradation, including the preservation process, preservation agents, specimen
37
preparation, contamination or simply age (reviewed in (Doucet and Hill 2009). Given that museum skin specimens are widely used in studies of avian coloration (e.g., Bridge et al. 2008, Kennedy 2010), I suggest that skin specimen coloration should be pre-tested against live birds;
measurements obtained from skin samples should be corrected for age and/or condition of preservation, and the results should be interpreted with greater caution.
References
Alonso-Alvarez C, Doutrelant C, Sorci G (2004) Ultraviolet reflectance affects male-male interactions in the blue tit (Parus caeruleus ultramarinus). Behav Ecol 15:805-809.
Amundsen T, Forsgren E (2001) Male mate choice selects for female coloration in a fish. P Natl Acad Sci USA 98:13155-13160.
Andersson M B (1994) Sexual selection. Princeton University Press, NJ.
Armenta J K, Dunn P O, Whittingham L A (2008) Effects of specimen age on plumage color. Auk 125:803-808.
Balakrishnan P (2010) Reproductive biology of the square-tailed Black Bulbul Hypsipetes ganeesa in the Western Ghats, India. Indian Birds 5:134-138.
Bennett A T, Cuthill I C, Partridge J C, Lunau K (1997) Ultraviolet plumage colors predict mate preferences in starlings. P Natl Acad Sci USA 94:8618-8621.
Borgia G (1995) Complex male display and female choice in the spotted
Borgia G (1995) Complex male display and female choice in the spotted