Biological Journal of the Linnean Society
Sperm size evolution in African greenbuls (Passeriformes: Pycnonotidae)
Journal:For Biological Peer Journal of theReview Linnean Society Manuscript ID: BJLS 4087.R1
Manuscript Type: Research Article
Date Submitted by the Author: n/a
Complete List of Authors: Omotoriogun, Taiwo; University of Oslo, Natural History Museum Albrecht, Tomas; Charles University in Prague, Department of Zoology; Academy of Sciences of the Czech Republic, Institute of Vertebrate Biology Hořák, David; Charles University in Prague, Department of Ecology Laskemoen, Terje; University of Oslo, Natural History Museum Ottosson, Ulf; University of Jos, A.P. Leventis Ornithological Research Institute Rowe, Melissah; University of Oslo, Natural History Museum Sedlacek, O; Charles University in Prague, Faculty of Science, Department of Ecology ; Lifjeld, Jan; University of Oslo, Natural History Museum
sperm competition, diversification, phylogenetic signal, evolutionary rate, Keywords: Pycnonotidae
Biological Journal of the Linnean Society Page 1 of 40 Biological Journal of the Linnean Society
1 2 3 1 Sperm size evolution in African greenbuls (Passeriformes: Pycnonotidae) 4 5 2 6 7 3 TAIWO C. OMOTORIOGUN, 1, 2 * TOMAS ALBRECHT, 3,4 DAVID HOŘÁK, 5 TERJE 8 9 LASKEMOEN, 1 ULF OTTOSSON, 2 MELISSAH ROWE, 1,6 ONDŘEJ SEDLÁČEK 5 and JAN T. 10 4 11 1 12 5 LIFJELD 13 14 6
15 1 16 7 Natural History Museum, University of Oslo, Oslo, Norway
17 2 18 8 A.P. Leventis OrnithologicalFor Research Peer Institute, Un Reviewiversity of Jos, Nigeria 19 3 20 9 Institute of Vertebrate Biology, Academy of Sciences of the Czech Republic, Brno, Czech Republic 21 4 22 10 Department of Zoology, Faculty of Science, Charles University in Prague, Prague, Czech Republic 23 5 24 11 Department of Ecology, Faculty of Science, Charles University in Prague, Prague, Czech Republic 25 6 26 12 Centre for Ecological and Evolutionary Synthesis, Department of Biosciences, University of Oslo, 27 28 13 Oslo, Norway 29 30 14 31 32 15 *Corresponding author. 33 34 16 Email: [email protected] 35 36 17 Telephone: +4745529529 37 38 18 39 40 19 Running title: Sperm size evolution in African greenbuls 41 42 20 43 44 21 45 46 22 47 48 23 49 50 24 51 52 25 53 54 26 55 56 27 57 58 59 60 1 Biological Journal of the Linnean Society Biological Journal of the Linnean Society Page 2 of 40
1 2 3 28 ABSTRACT 4 5 29 Sperm morphology is highly diversified across the animal kingdom and recent comparative evidence 6 7 30 from passerine birds suggests that postcopulatory sexual selection is a significant driver of sperm 8 9 31 evolution. Here, we describe sperm size variation among 20 species of African greenbuls and one 10 11 32 bulbul (Passeriformes: Pycnonotidae), and analyse the evolutionary differentiation of sperm size 12 13 33 within a phylogenetic framework. We found significant inter specific variation in sperm size; with 14 15 34 some genera exhibiting relatively long sperm (e.g. Eurillas ) and others short sperm head lengths (e.g. 16 17 35 Phyllastrephus ). However, our results suggest that contemporary levels of sperm competition are 18 For Peer Review 19 36 unlikely to explain sperm diversification within this clade: the coefficients of inter male variation 20 21 37 (CV bm ) in sperm length were generally high, suggesting relatively low and homogeneous rates of 22 23 38 extra pair paternity. Finally, in a comparison of six evolutionary or tree transformation models, we 24 25 39 found support for both the Kappa (evolutionary change primarily at nodes) and Lambda (lineage 26 27 40 specific evolutionary rates along branches) models in the evolutionary trajectories of sperm size 28 29 41 among species. We therefore conclude that African greenbuls have more variable rates of sperm size 30 31 42 evolution than expected from a neutral model of genetic drift. Understanding the evolutionary 32 33 43 dynamics of sperm diversification remains a future challenge. 34 35 44 36 37 45 KEYWORDS: sperm competition, diversification, phylogenetic signal, evolutionary rate, 38 39 Pycnonotidae. 40 46 41 42 47 43 44 48 45 46 49 47 48 50 49 50 51 51 52 52 53
54 53 55 54 56 57 55 58 59 60 2 Biological Journal of the Linnean Society Page 3 of 40 Biological Journal of the Linnean Society
1 2 3 56 INTRODUCTION 4 5 57 Across the animal kingdom, sperm cells are highly diversified in size, shape and structure (Cohen, 6 7 58 1977; Pitnick, Hosken & Birkhead, 2009). There is a strong phylogenetic signal in this diversity, such 8 9 59 that sperm traits can be informative in systematics and taxonomy (Jamieson, Ausio & Justin . 1995). 10 11 60 Nevertheless, it remains unclear why sperm cells have diversified to such a great extent given their 12 13 61 common function of locating and fertilizing ova. It is presumed that this diversity reflects the outcome 14 15 62 of genetic drift over evolutionary time scales, or is driven by selection. Sperm must perform in an 16 17 63 environment that can exert various selection pressures on them. For birds, which are internal 18 For Peer Review 19 64 fertilizers, this environment is the female oviduct. Here, sperm need to cross various biochemical, 20 21 65 physiological, morphological and behavioural barriers to their successful insemination, storage, 22 23 66 migration and eventually fertilization of the egg (Birkhead et al. 1993; Pitnick et al. 2009). These 24 25 67 challenges put forth by the female reproductive tract can vary across species, as can the level of sperm 26 27 68 competition. Sperm competition arises when sperm from two or more males compete for fertilization 28 29 69 of the same ova (Parker, 1970). Differences in sperm competitiveness among males can therefore 30 31 70 create the opportunity for postcopulatory sexual selection, which may lead to evolutionary changes in 32 33 71 sperm traits. Moreover, there is a theoretical possibility for female mate preferences in postcopulatory 34 35 72 sexual selection (i.e. cryptic female choice, Eberhard, 1996; Snook, 2005). One possible way to look 36 37 73 for signatures of selection is to perform comparative analyses of sperm differentiation within a 38 39 phylogenetic framework. If sperm evolve purely by random drift (Brownian motion), divergences 40 74 41 42 75 between taxa or lineages are expected to be proportional to the phylogenetic distance between them 43 44 76 (Pagel, 1997 ; Blomberg1, Garland & Ives, 2003). Deviations from such a covariance pattern might 45 46 77 suggest variable rates of evolutionary change, either among lineages or for different time periods in 47 48 78 the evolutionary history of a group. Here we apply this approach to the study of sperm length 49 50 79 evolution in a group of passerine birds with a well resolved, time calibrated phylogeny. 51 52 80 The order Passeriformes is the largest avian order, encompassing a majority of all extant 53 54 81 species (Gill & Donsker, 2015). Passerine birds have a unique sperm morphology characterized by an 55 56 82 enlarged and pointed acrosome on a helically shaped head and an elongated midpiece coiled around 57 58 83 the flagellum to form a mitochondrial helix (Humphreys, 1972; Koehler, 1995; Jamieson, 2006). 59 60 3 Biological Journal of the Linnean Society Biological Journal of the Linnean Society Page 4 of 40
1 2 3 84 Flagellum length appears considerably more variable among passerines relative to any other avian 4 5 85 order, and especially so within the Passerida parvorder ( sensu Sibley & Ahlquist 1990) of oscine 6 7 86 songbirds (Jamieson, 2006). Here, members of each of the three larger superfamilies Sylvioidea, 8 9 87 Muscicapoidea and Passeroidea, display the maximum range of interspecific sperm length variation 10 11 88 known for birds, roughly 40 µm to 300 µm (see Pitnick et al . 2009; Lifjeld et al . 2010; Immler et al . 12 13 89 2011, 2012 for lists of species specific sperm lengths). Passerines also appear to have higher levels of 14 15 90 sperm competition relative to the other avian orders, though there is still considerable variation among 16 17 91 species (Westneat & Sherman, 1997; Griffith, Owens & Thuman, 2002). Recent comparative studies 18 For Peer Review 19 92 have revealed three general patterns that link sperm length variation to the level of sperm competition 20 21 93 in passerines. 22 23 94 First, there is general trend that longer sperm have evolved in taxa with high sperm 24 25 95 competition (Briskie, Montgomerie and Birkhead, 1997; Kleven et al . 2009, Immler et al . 2011). A 26 27 96 similar pattern is observed for other animal groups, including insects (Morrow & Gage, 2000), fish 28 29 97 (Balshine et al . 2001) and mammals (Gomendio & Roldan, 1991; Tourmente, Gomendio & Roldan, 30 31 98 2011; but see Gage & Freckleton, 2003). In birds, however, the relationship does not appear to be 32 33 99 linear and there are many species with high sperm competition that exhibit relatively short sperm 34 35 100 (Immler & Birkhead, 2007; Immler et al . 2011). Second, pairs of closely related species with high 36 37 101 sperm competition have more divergent sperm lengths than those with low sperm competition (Rowe 38 39 et al . 2015). This indicates that the rate of evolutionary change in sperm length is higher in species 40 102 41 42 103 with more sperm competition, and also suggests that changes may go in either direction and not 43 44 104 always towards longer sperm. Finally, there is a strong negative association between the level of 45 46 105 sperm competition and the variation in sperm length among males in a population (Calhim, Immler & 47 48 106 Birkhead , 2007; Kleven et al . 2008; Lifjeld et al . 2010; Laskemoen et al . 2013). This is consistent 49 50 107 with a model of stabilizing selection where males with sperm sizes around the population mean are 51 52 108 predicted to be more successful in sperm competition. Thus, sperm competition seems to be a strong 53 54 109 force of stabilizing selection, which over evolutionary time scales causes rapid evolution and 55 56 110 diversification in sperm length. Stabilizing selection causing trait divergence may seem paradoxical, 57 58 59 60 4 Biological Journal of the Linnean Society Page 5 of 40 Biological Journal of the Linnean Society
1 2 3 111 but it is not. Stabilizing selection with a moving adaptive peak is a well recognized process of 4 5 112 evolutionary change (Estes & Arnold, 2007). 6 7 113 Here, we analysed variation in sperm length among 20 species of African greenbuls and one 8 9 114 species of bulbul, all belonging to the Pycnonotidae family, which is part of the Sylvioidea clade with 10 11 115 larks, swallows, and several families of warblers and babblers as their closest relatives (Fregin et al . 12 13 116 2012). The Pycnonotidae consists of two major clades; the African greenbul radiation and the Asian 14 15 117 bulbul radiation ( Pasquet et al . 2001; Moyle & Marks, 2006). The African bulbuls (Pycnonotus ) 16 17 118 belong to the Asian radiation and have more recently colonized Africa. The African greenbuls consist 18 For Peer Review 19 119 of about 60 species from 13 genera (Gill & Donsker, 2015) . The phylogeny of the group is now well 20 21 120 resolved and the revised classification reflects monophyletic genera (Johansson et al . 2007; Jetz et al . 22 23 121 2012). Our study species represent six genera of greenbuls from Western Africa, for which there is 24 25 122 almost no information available concerning sperm morphology; as is indeed the case for most African 26 27 123 birds. 28 29 124 Our main aim was to examine how sperm size has diversified over the evolutionary history of 30 31 125 our study species and to test how well various evolutionary models might explain the contemporary 32 33 126 interspecific variation in sperm total length and length of sperm components (i.e. head, midpiece and 34 35 127 flagellum lengths). We mapped species’ sperm lengths onto an ultrametric tree constructed from the 36 37 128 most comprehensive multilocus phylogenies available (Jetz et al . 2012), supplemented with some of 38 39 our own sequences of a mitochondrial gene, and tested the fit of a range of evolutionary or tree 40 129 41 42 130 transformation models. We also quantified intraspecific variation in sperm total length as a proxy for 43 44 131 extra pair paternity, in order to test for a possible signal of sperm competition in the diversification of 45 46 132 sperm size. 47 48 133 49 50 134 MATERIAL AND METHODS 51 52 135 STUDY SPECIES 53 54 136 African greenbuls are characteristically cryptic, olive green to brown, medium sized (c. 13–26 cm) 55 56 137 birds occurring in the understory and canopies of Afrotropical forests. They are largely frugivorous. 57 58 138 The sexes show plumage monomorphism, while size dimorphism exists in some species and in these 59 60 5 Biological Journal of the Linnean Society Biological Journal of the Linnean Society Page 6 of 40
1 2 3 139 instances males are larger than females (Keith, Urban & Fry, 1992 ). The mating system is 4 5 140 predominantly monogamy (Fry, Keith & Urban, 2000), with the exception of Eurillas latirostris , 6 7 141 which has been classified as a lekking species (Brosset, 1982). We collected data from six greenbul 8 9 142 genera: Eurillas (five species), Phyllastrephus (six species), Criniger (three species), Bleda (three 10 11 143 species), Arizelocichla (two species), and Chlorocichla (one species). These species are mainly 12 13 144 distributed in the lowland rainforest, but Phyllastrephus and Arizocichla greenbuls occur in montane 14 15 145 forests where they seem to have radiated quite recently (Fjeldså et al. 2007). In addition to the 20 16 17 146 species of greenbul, we included one species of bulbul, Pycnonotus barbatus. This species is common 18 For Peer Review 19 147 and widely distributed in various habitats in Africa. 20 21 148 22 23 149 DATA COLLECTION AND SAMPLING PROCEDURE 24 25 150 We captured birds using mist nets during the breeding season in 2010 to 2013 in Nigeria and 26 27 151 Cameroon. Sampling in Nigeria was conducted at a range of sites, including Amurum Forest Reserve, 28 29 152 Jos (09 °53′ N, 08 °59′ E), Omo Forest Reserve, Ogun (06 °51′ N, 4 °30′ E), International Institute of 30 31 ° ° ° 32 153 Tropical Agriculture (IITA), Ibadan (07 30′ N, 03 55′ E), and Okomu National Park, Benin (06 33′ N, 33 34 154 05 °26′E). In Cameroon, birds were sampled in the vicinity of Laide Farm, Bamenda Banso Highlands 35 36 155 (06 °05′ N, 10 °28′ E) and in Mt Cameroon National Park (04 °15′ N, 09 °09′E). 37 38 156 Sperm samples were obtained by cloacal massage (Wolfson, 1952), whereby the exuded 39 40 157 semen of 0.5–3 µl was collected by a 10 µl capillary tube, diluted in a small volume (c. 20 µl) of 41 42 158 phosphate buffered saline and then fixed in 300 µl of 5% formaldehyde solution for later slide 43 44 159 preparation. We also collected a blood sample from the brachial vein for DNA extraction and 45 46 160 sequencing of the mitochondrial cytochrome oxidase I (COI) gene as part of an ongoing effort to 47 48 161 build a DNA barcode library for West African birds (cf. Hebert, Ratnasingham & deWaard, 2003). 49 50 162 Birds were fitted with uniquely numbered aluminium band (from SAFRING) to prevent resampling of 51 52 163 individuals. For all our study species, body mass information was taken from Fry et al . (2000). 53 54 164 55 56 165 SPERM MORPHOLOGY 57 58 59 60 6 Biological Journal of the Linnean Society Page 7 of 40 Biological Journal of the Linnean Society
1 2 3 166 A small aliquot ( c. 15 µl) from each formaldehyde fixed sperm sample was applied onto a microscope 4 5 167 glass slide and allowed to air dry. We then gently rinsed slides with distilled water and air dried them 6 7 168 again. Next, high magnification (160× or 320×) digital images of sperm cells were taken using a Leica 8 9 169 DFC420 camera mounted on a Leica DM6000 B digital light microscope (Leica Microsystem, 10 11 170 Heerbruug, Switzerland). The Leica Application Suite (version 2.6.0 R1) was used to measure (± 0.1 12 13 171 µm) the length of the sperm head, midpiece and tail (i.e. the section of the flagellum not entwined by 14 15 172 the midpiece), from which we calculated flagellum length (sum of midpiece and tail length), sperm 16 17 173 total length (sum of head and midpiece and tail length) and the ratios of midpiece:flagellum length, 18 For Peer Review 19 174 flagellum:head length and midpiece:sperm total length. We measured 10 morphologically intact 20 21 175 spermatozoa for each male (i.e. no head damage or broken tail) following the recommendation in 22 23 176 Laskemoen et al. (2007). Sperm measurements were highly repeatable for head, midpiece and tail (all 24 25 177 r> 80%, all P< 0.001). 26 27 178 We calculated the coefficient of intra male (CV wm ) and inter male (CV bm ) variation in sperm 28 29 179 total length using the formula, CV = (SD/Mean) × 100. For the CV metric, we corrected for sample 30 bm 31 × 32 180 size ( n) variation using CV bm = (SD/Mean) 100 (1 + (1/4n)), as recommended in Sokal & Rohlf 33 34 181 (1981). 35 36 182 37 38 183 SPECIES PHYLOGENY 39 40 184 The phylogeny for our study species was obtained from www.birdtree.org (Jetz et al . 2012), which 41 42 185 comprises publicly available molecular sequence data for a wide range of avian species. We 43 44 186 downloaded 1000 phylogenetic trees (Hackett backbone) for 18 of our 21 study species and 45 46 187 summarised these trees onto a single maximum clade credibility tree using median node heights at 0.5 47 48 188 posterior probability limits in TreeAnnotator (version 1.6.2, Rambaut & Drummond, 2009). We then 49 50 189 manually coded the three remaining species (i.e. those with missing sequence data from Jetz et al . 51 52 190 (2012)) into the maximum clade credibility tree (i.e. at the middle branch length of their sister taxon) 53 54 191 based on literature sources for Phyllastrephus poliocephalus (Zuccon & Ericson, 2010) and a 55 56 192 mitochondrial gene tree (COI) derived from our study individuals for both Phyllastrephus baumanni 57 58 193 and Chlorocichla simplex . 59 60 7 Biological Journal of the Linnean Society Biological Journal of the Linnean Society Page 8 of 40
1 2 3 194 To obtain this COI tree, we sequenced the first part of the COI gene, between 650 and 750 bp 4 5 195 in length corresponding to the standard DNA barcode marker for animals (Hebert et al . 2003; see 6 7 196 Appendix S1 and Table S1in Supplementary Information). Sequences are available in the folder 8 9 197 BONGR at the BOLD database (Ratnasingham & Hebert, 2007). We first aligned sequences using 10 11 198 ClustalW in the program MEGA version 6.06 (Tamura et al . 2013), and then applied the Kimura 2 12 13 199 parameter model to construct a Maximum Likelihood tree with branch length at 10000 bootstrap 14 15 200 iterations. Species nomenclature is based on the IOC World Bird List (Gill & Donsker, 2015). 16 17 201 18 For Peer Review 19 202 DATA ANALYSIS 20 21 203 All analyses were performed with the statistical package R (version 2.15.2, R Development Core 22 23 204 Team, 2013). We applied log transformations to improve distributions for all sperm traits prior to 24 25 205 analysis, with the exception of the ratios of sperm midpiece:flagellum length and sperm midpiece:total 26 27 206 length which were logit transformed following the recommendation of Warton & Hui (2011). We used 28 29 207 an ANOVA to test for differences in sperm traits (i.e. the length of sperm head, midpiece, flagellum, 30 31 208 total length and CV ) among species, and tested for differences in the CV of total sperm length 32 wm bm 33 209 using homogeneity of variance tests (Levene’s test). 34 35 210 We performed phylogenetic generalized least squares (PGLS) regressions to examine 36 37 211 associations among sperm traits, and to test whether sperm size was associated with male body mass. 38 39 Separate models were run for each sperm trait. The PGLS approach accounts for the statistical non 40 212 41 42 213 independence of data points due to common ancestry of species (Pagel, 1999; Freckleton, Harvey & 43 44 214 Pagel, 2002) and allows the estimation (via maximum likelihood) of the phylogenetic scaling 45 46 215 parameter lambda ( λ): λ values = 0 indicate phylogenetic independence, while values = 1 indicate 47 48 216 phylogenetic dependence. We tested the likelihood ratio of λ value against λ = 1 and λ = 0. PGLS 49 50 217 regressions were performed using package ‘caper’ (Orme et al . 2012 ). 51 52 218 To quantify the phylogenetic signal in sperm traits, we calculated Pagel’s λ (Pagel, 1999) and 53 54 219 Blomberg’s K (Blomberg et al . 2003) using the package ‘phytools’ (Revell, 2012). Log likelihood 55 56 220 ratio tests were used to determine if estimated maximum likelihood values for λ differed from 0 (i.e. 57 58 221 no phylogenetic signal), whereas for Blomberg’s K we used the randomization test to determine 59 60 8 Biological Journal of the Linnean Society Page 9 of 40 Biological Journal of the Linnean Society
1 2 3 222 whether traits exhibited a phylogenetic signal (i.e. K> 0). Values of K can exceed 1, in which case they 4 5 223 indicate more similarity among related taxa than expected under a Brownian motion model of trait 6 7 224 evolution. We used these two measures (i.e. Pagel’s λ and Blomberg’s K) as they are not identical 8 9 225 measures of phylogenetic signal; rather λ measures the strength of the phenotypic – genotypic 10 11 226 covariance assuming Brownian motion (λ = 1 equals Brownian motion), while K reflects the 12 13 227 partitioning of trait variance among and within clades: high K implies more variance among clades (i.e. 14 15 228 deeper in the phylogeny), whereas low K means more variance among the terminal branches. In 16 17 229 addition, we mapped sperm size evolution on the phylogeny using the contMap function in ‘phytools’ 18 For Peer Review 19 230 (Revell, 2013). This method allows for the visualisation of contemporary trait values as well as their 20 21 231 constructed phenotypic values at internal nodes in the tree. We visualised trait variation for both sperm 22 23 232 total length and sperm head length separately because of the different evolutionary trajectories of these 24 25 233 traits (Immler et al . 2011; Rowe et al . 2015). Additionally, we visualised ancestral trait values for 26 27 234 sperm total length using a traitgram using the function ‘phenogram’, and then extended this to 28 29 235 incorporate uncertainty in the reconstructed ancestral trait values using the function fancyTree in the 30 31 236 ‘phytools’ package (Revell, 2013). 32 33 237 Finally, we used the fitContinuous function in the ‘geiger’ package (Harmon et al. 2008) to 34 35 238 compare the fit of five tree transformation models against a null model of Brownian motion (BM), i.e. 36 37 239 sperm divergence is perfectly predicted by the phylogenetic distance. The models were 1) Lambda: 38 39 phenotypic divergence covaries with phylogenetic distance, but allows for variable evolutionary rates, 40 240 41 42 241 2) Delta: the evolutionary rate accelerates or decelerates over time, 3) Kappa: evolutionary change 43 44 242 occurs at speciation events, but is not proportional to branch length, 4) Ornstein Uhlenbeck (OU): a 45 46 243 random walk within a constrained trait space, where traits tend to converge towards a single value, and 47 48 244 5) Early Burst (EB): an early burst of trait diversification followed by reduced evolutionary rates (or
49 2 50 245 stasis). These models provide an estimation of the net rate of evolution (σ ) for the trait in question. 51 52 246 For models departing from a simple BM process, a number of additional parameters that describe the 53 54 247 evolutionary trajectory of a trait are also estimated. The Lambda model estimates the parameter λ, 55 56 248 which describes the extent to which phylogeny predicts covariance among trait for species. The Delta 57 58 249 model estimates the parameter δ, which compares the contributions of early versus late evolution 59 60 9 Biological Journal of the Linnean Society Biological Journal of the Linnean Society Page 10 of 40
1 2 3 250 across a phylogeny; δ = 1 indicates gradual evolution, 0 < δ < 1indicates most trait evolution is near 4 5 251 the base of the tree, whereas δ > 1 indicates most trait evolution occurs near the tips of the tree. The 6 7 252 Kappa model estimates the parameter κ, where κ = 1 indicate gradual evolution across the phylogeny, 8 9 253 κ = 0 implies a punctuated model of evolution with evolutionary change associated with speciation 10 11 254 events, 0 < κ < 1indicates more trait evolution than expected on shorter branches and thus more stasis 12 13 255 on longer branches, while κ >1 indicates more trait evolution than expected on longer branches. The 14 15 256 Ornstein Uhlenbeck model includes the parameter α, which reflects the evolutionary constraint on 16 17 257 trait evolution or the ‘attraction’ towards a single optimal phenotypic value, and as α approaches 0 the 18 For Peer Review 19 258 model collapses to a BM model. Finally, in the Early Burst model, the additional parameter is r, which 20 21 259 indicates the change in rate of trait evolution through time; when r = 0 the model collapses to a pure 22 23 260 BM model in which σ 2 is constant. 24 25 261 We compared models using the Akaike Information Criterion corrected for small sample size 26 27 262 (AICc); the model with the lowest AICc value indicates the best fit model. We also calculated Akaike 28 29 30 263 weights for all models and used both AICc and Akaike weights values to assess model support. 31 ≤ 32 264 Value of AICc 2 indicates substantially supported models, while those in which 4 ≤ AICc ≤ 7 33 34 265 indicates less plausible models (Burnham & Anderson, 2004). We analysed the evolution of head 35 36 266 length, midpiece length, flagellum length and total sperm length separately. 37 38 267 39 40 268 RESULTS 41 42 269 Sperm total length ranged from 70 µm in Phyllastrephus baumanni to 117 µm in Eurillas curvirostris 43 44 270 (Table 1; Table S2). All sperm traits showed significant variation among species (Table 1), though 45 46 271 values for sperm head length varied within a narrow range (11–16 µm). In contrast, sperm total length 47 48 272 was highly variable and most of this variation was explained by the length of the flagellum (Table 1; 49 50 273 Fig. 1). The coiled midpiece was typically elongated and extended two thirds or more along the length 51 52 274 of the flagellum (Fig. 1). 53 54 275 Sperm head length showed significant negative association with midpiece length (β = –0.03 ± 55 56 276 0.01 SE, t = –5.24, P<0.001, λ = 1 0.005; 1.00 ) but not with flagellum length ( β = 0.02 ± 0.02SE, t = 1.22, 57 58 59 60 10 Biological Journal of the Linnean Society Page 11 of 40 Biological Journal of the Linnean Society
1 2 0.002; 0.006 3 277 P = 0.24, λ = 0.80 ) among species (see PGLS regression among sperm traits; Table S3). Sperm 4 5 278 total length was not significantly associated with male body mass ( β = –0.54 ± 0.39 SE, t = –1.37, P = 6 7 279 0.19, λ = 0.70 0.05; <0.001 ). There was significant heterogeneity of variances among species for sperm 8 9 280 midpiece length (Levene’s test: F20, 145 = 1.80, P = 0.03) but not for flagellum length ( F20, 145 = 1.08, P 10 11 281 = 0.38) or total sperm length (Levene’s test: F20, 145 = 0.91, P = 0.57). The homogeneity of variances 12 13 282 for total sperm length implies that the corresponding coefficients of variation in male sperm lengths 14
15 283 (i.e. CV bm ; Table 1) did not vary significantly among species. As the sperm length CV bm metric is 16 17 284 negatively correlated with the rate of extra pair paternity in passerine birds (Calhim et al . 2007; 18 For Peer Review 19 285 Immler et al . 2008; Kleven et al. 2008; Lifjeld et al . 2010), these results suggest that there is little or 20 21 286 no variation among the study species in extra pair paternity. The average sperm total length CV bm 22 23 287 value for the 12 species for which sperm length was measured for >3 males, was 2.82 ± 0.89 SD 24 25 288 (range 1.61–4.23; Table 1). 26 27 289 Mapping sperm total length onto the phylogenetic tree (Figs 2A; 3), we found that the majority 28 29 290 of species (N = 12), with representatives from all genera except Eurillas , exhibited total sperm length 30 31 291 within a relatively narrow range of 79 µm to 89 µm, which is close to the estimated ancestral value for 32 33 292 sperm total length (84 µm) for the group (Fig. 3). The Eurillas had consistently longer sperm (103– 34 35 293 117 µm) than all other genera. Within this genus, the sister species Eurillas ansorgei and E. gracilis 36 37 294 seem to have diverged fairly rapidly in total sperm length (Fig. 2A; Table 1). The genus 38 39 Phyllastrephus is characterized by a short sperm head (Fig. 2B); values ranged from 11.8 µm to12.5 40 295 41 42 296 µm, which was not overlapping with the other genera (13.4–15.5 µm; Table 1). In three genera, 43 44 297 Phyllastrephu s, Criniger , Arizelocichla , single species have evolved considerably shorter sperm total 45 46 298 lengths than their congeners, i.e. around 70 µm. Finally, the genus Criniger appeared to show rapid 47 48 299 divergence in sperm total length (Figs 2A; 3; Table 1), especially in the sister species Criniger 49 50 300 barbatus and Criniger chloronotus . Sample sizes were admittedly quite low, but assuming that their 51 52 301 intraspecific variances in sperm length are similar to those of the other greenbuls; the data do suggest 53 54 302 this clade may have undergone very rapid sperm evolution. 55 56 303 When we tested for a phylogenetic signal in the sperm sizes, we found an interesting contrast 57 58 304 between the results for Pagel’s λ and Blomberg’s K (Table 2). Pagel’s λ indicated a significant 59 60 11 Biological Journal of the Linnean Society Biological Journal of the Linnean Society Page 12 of 40
1 2 3 305 phylogenetic signal for all traits, except midpiece length, which showed no significance ( P = 0.203). 4 5 306 However, all Blomberg’s K values were low and non significant for all traits. Since Blomberg’s K is 6 7 307 sensitive to variation among terminal branches, the putative rapid divergence between the two 8 9 308 Criniger sister species may have had a large influence on signal strength in our dataset. When we 10 11 309 removed Criniger barbatus from the test, values for Blomberg’s K exceed 1.3 and revealed a 12 13 310 significant phylogenetic signal (likelihood ratio test, all P<0.003) for all sperm component lengths and 14 15 311 their ratios. 16 17 312 The tests of five different models for sperm traits’ evolution suggest that evolutionary 18 For Peer Review 19 313 trajectories in sperm total length, flagellum and head length were often best explained by the Kappa 20 21 314 model ( AICc = 0; Table 3), For flagellum length and sperm total length the Lambda model also had 22 23 315 reasonable support (Table 3). For midpiece length, the Lambda model had the best support but the 24 25 316 Kappa and OU models also had reasonable support (Table 3). Finally, the evolutionary trajectory of 26 27 317 sperm head length was best explained by the Kappa model. The other models (BM, Delta and EB) 28 29 318 assume that the evolutionary rate changes over time within lineages in various ways, and they all 30 31 received no support for the evolution of sperm traits. 32 319 33 34 320 35 36 321 DISCUSSION 37 38 322 Here, we analysed sperm size diversification in a group of endemic African passerines the greenbuls. 39 40 323 Very little information exists on sperm morphology for this group (two species included in Albrecht et 41 42 324 al . 2013, Table S1), thus the descriptive data on sperm morphology presented here contribute to the 43 44 325 general knowledge base for the individual species, and also fill a gap in our broader understanding of 45 46 326 how sperm morphology varies among clades in the passerine phylogeny. More importantly, through 47 48 327 the use of the analysis of evolutionary trajectories of sperm size diversification in a phylogenetic time 49 50 328 calibrated framework the results indicate lineage specific rates of sperm evolution in this group. The 51 52 329 diversity of sperm sizes among contemporary species therefore appears to not only be a result of 53 54 330 neutral evolution by genetic drift, but suggests a role for selection and constraints. In the following we 55 56 331 discuss these perspectives in more detail. 57 58 59 60 12 Biological Journal of the Linnean Society Page 13 of 40 Biological Journal of the Linnean Society
1 2 3 332 Afrotropical birds are less well studied than birds in other regions of the world, particularly the 4 5 333 temperate zones (see Macedo, Karubian & Webster, 2008; Reddy, 2014). This general pattern also 6 7 334 holds true for descriptive data on sperm morphology. Our study confirms that the African greenbuls 8 9 335 exhibit the typical filiform passerine sperm with a corkscrew shaped head and an elongated midpiece 10 11 336 consisting of a mitochondrial helix coiled around most of the flagellum. An extended midpiece along a 12 13 337 very long flagellum is typically seen in the Passerida group of the oscine passerines (Jamieson, 2006), 14 15 338 to which the greenbuls belong. Within this group, sperm sizes for certain species can reach nearly 300 16 17 339 µm. In the Hirundinidae family, which is closely related to the Pycnonotidae (Fregin et al . 20102), 18 For Peer Review 19 340 sperm lengths can reach up to 240 µm, as exemplified by the tree swallow Tachycineta bicolor 20 21 341 (Laskemoen et al . 2010, Immler et al . 2011). However, greenbul sperm are much shorter than this and 22 23 342 lie within a relatively narrow range of 70 µm to 120 µm. This is a quite common size range for many 24 25 343 Passeridan taxa, including several families within the Sylvioidea superfamily that are closely related to 26 27 344 the Pycnonotidae, like Old World warblers, Sylviid babblers, larks and long tailed tits (cf. sperm 28 29 345 lengths for species listed in Lifjeld et al . 2010, Immler et al . 2011, Immler, Gonzalez Voyer & 30 31 346 Birkhead, 2012). Thus, the sperm of African greenbuls are of similar size as their closest relatives, and 32 33 347 they share the general pattern of a significant size variation among species. 34 35 348 Our results also show that the variance in sperm lengths among males in a population is rather 36 37 349 homogeneous across species. Because sperm length variance (CV ) is negatively related to the 38 bm 39 frequency of extra pair paternity (Calhim et al . 2007, Lifjeld et al . 2010, Laskemoen et al . 2013), the 40 350 41 42 351 homogeneous variances suggests that the level of sperm competition is not especially variable among 43 44 352 our greenbul species. The average CV bm value for the group (2.87) gives an estimate of 14% extra pair 45 46 353 young when applying the formula given in Lifjeld et al . (2010, Fig. 2), which is a quite moderate level 47 48 354 for passerine birds (Griffiths et al . 2002). As far as we are aware, there are no published paternity 49 50 355 studies from the Pycnonotidae. The lack of support for inter specific variation in the level of extra pair 51 52 356 paternity suggests there is little to no scope for detecting signatures of sperm competition in the 53 54 357 evolution of sperm traits in this group (see PGLS regression between CV bm and sperm traits, Table S5). 55 56 358 The comparative analyses of sperm diversification within the greenbul phylogeny revealed a 57 58 359 clear signature of phylogeny, in which the magnitude of divergence between any two lineages is 59 60 13 Biological Journal of the Linnean Society Biological Journal of the Linnean Society Page 14 of 40
1 2 3 360 significantly influenced by the time since they split. However, this pattern was not consistent with a 4 5 361 Brownian motion model of neutral evolution, because lineages did not have a constant rate of sperm 6 7 362 evolution. There are several examples of variable divergence rates in the traitgram (Fig. 3), where 8 9 363 single species or lineages (e.g. Eurillas ) rapidly diverge from their relatives. These rapid divergences 10 11 364 occurred for some lineages early in the evolutionary history of the group, as shown by the early 12 13 365 increase in sperm length for the Eurillas greenbuls. Single species within the Arizelocichla , Criniger 14 15 366 and Phyllastrephus diverged from their congeners at the mid age of the phylogeny and evolved shorter 16 17 367 sperm. There is also a striking example of a recent and seemingly rapid divergence in sperm length for 18 For Peer Review 19 368 the closely related Criniger barbatus and Criniger chloronotus , which in some earlier classifications 20 21 369 (e.g. Howard & Moore, 2 nd edition, 1991), were considered conspecific subspecies. Taken together, 22 23 370 these divergences leave a clear impression that sperm size can evolve fast in some lineages and be 24 25 371 rather stable in others at a given point in time in the phylogeny. 26 27 372 We found that evolutionary diversification in sperm size in this group was best supported by 28 29 373 the Kappa model which suggests that most divergence in sperm size occurred shortly after the 30 31 374 speciation event (the nodes) and evolution was proportionally faster in shorter branches, so more stasis 32 33 375 on longer branches. We also found reasonable support for the Lambda model in the evolutionary 34 35 376 trajectories of sperm size. This model allows for variable rates of trait evolution among clades or 36 37 377 lineages. A constant rate of evolution among lineages would be identical to the Brownian motion 38 39 model. For the midpiece length, evolutionary trajectories were supported by multiple models: the 40 378 41 42 379 Lambda, Kappa and OU models received substantial support (all AICc <2). Generally, there was no 43 44 380 evidence that the diversification in sperm traits occurred predominantly early in the phylogeny (the 45 46 381 Early Burst model) and/or that sperm lengths accelerated or decelerated within lineages (Delta model). 47 48 382 Compared to the midpiece, flagellum and sperm total length where evolutionary trajectories 49 50 383 were supported by two or more models respectively, the evolution of sperm head length was only 51 52 384 supported by the Kappa model. Generally, there is a consistent pattern in passerine birds that shows 53 54 385 evolution of the sperm head differs fundamentally from the evolution of the flagellum or sperm total 55 56 386 length (Immler et al . 2011; Rowe et al . 2015). For example, in a recent comparative analysis, Rowe et 57 58 387 al . (2015) showed that sperm competition had a significant effect on the divergence rates in sperm 59 60 14 Biological Journal of the Linnean Society Page 15 of 40 Biological Journal of the Linnean Society
1 2 3 388 total length for 36 pairs of passerine species, whereas sperm competition had no such influence on the 4 5 389 divergence in sperm head length. The results also suggested sperm head size is evolutionarily 6 7 390 constrained, whereas there was no evidence for such constraints on midpiece, flagellum or total sperm 8 9 391 length (Rowe et al . 2015). 10 11 392 The variable rate of sperm size evolution observed among greenbul species poses new 12 13 393 questions about the mechanisms behind sperm diversification. The support for the Kappa model of 14 15 394 evolution suggests that sperm size evolves particularly fast around speciation or splitting events. This 16 17 395 might be explained by sperm divergences being accelerated by postcopulatory sexual selection at the 18 For Peer Review 19 396 early stages of speciation, e.g. by reinforcement. Our sperm data suggest that greenbuls have a mating 20 21 397 system with sperm competition, although at moderate levels for passerine birds. However, sperm 22 23 398 competition does not seem to be much variable among species, at least not among our contemporary 24 25 399 study populations. Therefore, the variable rates of sperm evolution among lineages can hardly be 26 27 400 explained by different levels of sperm competition in these lineages. Thus we suggest that the variable 28 29 401 rates of sperm size evolution must have other explanations than sperm competition per se and that 30 31 402 determining what these factors might be remains a major challenge for future studies of sperm 32 33 403 evolution. 34 35 404 36 37 405 ACKNOWLEDGEMENTS 38 39 We thank the Research Council of Norway (project no. 196554/V40 to JTL), Norwegian State 40 406 41 42 407 Educational Loan Fund (PhD scholarship to TCO), Czech Science Foundation (project no. 14 36098G 43 44 408 to TA) and International Foundation for Science (grant no. TJ/32343 to TCO) for financial support. 45 46 409 Field work in Cameroon was covered by research permits issued by the Ministry of Research and 47 48 410 Innovations (nos. 2011 000079, 2012 000075/MINRESI/B00/C00/C10/nye), and export permits 49 50 411 were issued by the Ministry of Forest and Fauna (nos. 2013 1705, 2014 0104/PRS/MINFOF/SG/ 51 52 412 DFAP/SDVEF/SC. In Nigeria, the field work was conducted under the scheme of A.P. Leventis 53 54 413 Ornithological Research Institute, University of Jos, which also provided a vehicle for fieldwork. We 55 56 414 are also grateful to Deni Bown and Asiedu Robert who arranged for our access to the IITA forest, to 57 58 415 Constance Eno Crossby, Taiye Adeniyi Adeyanju, Silje Hogner and Lars Erik Johannessen for field 59 60 15 Biological Journal of the Linnean Society Biological Journal of the Linnean Society Page 16 of 40
1 2 3 416 assistance, and to Gunnhild Marthinsen for help in the laboratory. We thank two anonymous reviewers 4 5 417 for their useful comments. This is contribution no XX from the A. P. Leventis Ornithological Research 6 7 418 Institute. 8 9 419 10 11 420 REFERENCES 12 13 421 Albrecht T, Kleven O, Kreisinger J, Laskemoen T, Omotoriogun TC, Ottosson U, Reif J, Sedláček O, 14 422 Hořák D, Robertson RJ, Lifjeld JT. 2013. Sperm competition in tropical versus temperate zone birds. 15 423 Proceedings of the Royal Society of London Series B 280: 20122434. 16 424 Balshine S, Leach BJ, Neat F, Werner NY, Montgomerie R. 2001. Sperm size of African cichlids in relation 17 425 to sperm competition. Behavioral Ecology 12: 726–731. Birkhead TR, Møller AP, Sutherland WJ. 1993. Why do females make it so difficult for males to fertilize 18 426 427 their eggs? JournalFor of Theoretical Peer Biology 161: 51–60.Review 19 428 Blomberg SP, Garland T, Ives AR. 2003. Testing for phylogenetic signal in comparative data: Behavioral traits 20 429 are more labile. Evolution 57: 717–745. 21 430 Briskie JV, Montgomerie R, Birkhead TR. 1997. The evolution of sperm size in birds. Evolution 51: 937– 22 431 945. 23 432 Brosset A. 1982. The Social Life of the African Forest Yellow whiskered Greenbul Andropadus latirostris . 24 433 Zeitschrift für Tierpsychologie 60: 239–255. 25 434 Burnham KP, Anderson DR. 2004 . Multimodel inference: understanding AIC and BIC in model 26 435 selection. Sociological Methods and Research 33(2): 261–304. 27 436 Calhim S, Immler S, Birkhead TR. 2007. Postcopulatory sexual selection is associated with reduced variation 28 437 in sperm morphology. PLoS One 2: e413. 29 438 Cohen J. 1977. Reproduction . Butterworths, London. 30 439 Eberhard WG. 1996. Female control: sexual selection by cryptic female choice. Princeton, New Jersey: 440 Princeton University Press. 31 441 Estes S, Arnold SJ. 2007. Resolving the paradox of stasis: models with stabilizing selection explain 32 442 evolutionary divergence on all tmescales. American Naturalist 169: 227–244. 33 443 Fjeldså J, Johansson U, Lokugalappatti LGS, Bowie RK. 2007. Diversification of African greenbuls in space 34 444 and time: linking ecological and historical processes. Journal of Ornithology 148: 359–367. 35 445 Freckleton RP, Harvey PH, Pagel M. 2002. Phylogenetic analysis and comparative data: A test and review of 36 446 evidence. American Naturalist 160: 712–726. 37 447 Fregin S, Haase M, Olsson U, Alström P. 2012. New insights into family relationships within the avian 38 448 superfamily Sylvioidea (Passeriformes) based on seven molecular markers. BMC Evolutionary Biology 39 449 12: 157. 40 450 Fry CH, Keith S, Urban EK. 2000. The Birds of Africa. London: Academic Press. 41 451 Gage MJG, Freckleton RP. 2003. Relative testes size and sperm morphometry across mammals: no evidence 42 452 for an association between sperm competition and sperm length. Proceedings of the Royal Society of 43 453 London Series B 270: 625–632. 454 Gill F, Donsker D. 2015. IOC World Bird List version 4.2. Available at: www.worldbirdnames.org. 44 455 Gomendio M, Roldan ERS. 1991. Sperm competition influences sperm size in mammals. Proceedings of 45 456 the Royal Society of London Series B 243: 181–185. 46 457 Griffith SC, Owens IPF, Thuman KA. 2002. Extra pair paternity in birds: a review of interspecific 47 458 variation and adaptive function. Molecular Ecology 11: 2195–2212. 48 459 Harmon LJ, Weir JT, Brock CD, Glor RE, Challenger W. 2008. GEIGER: investigating 49 460 evolutionary radiations . Bioinformatics 24: 129–131. 50 461 Hebert PD, Ratnasingham S, deWaard JR. 2003 . Barcoding animal life: cytochrome c oxidase 51 462 subunit 1 divergences among closely related species. Proceedings of the Royal Society of 52 463 London Series B 270: S96–99. 53 464 Howard R, Moore A. 1991. A Complete Checklist of Birds of the World, Second Edition. 54 465 Humphreys PN. 1972. Brief observation on the semen and spermatozoa of certain passerine and non passerine 55 466 birds. Journal of Reproduction and fertility 29: 327–336. 56 467 Immler S, Birkhead T. 2007. Sperm competition and sperm midpiece size: no consistent pattern in passerine 57 468 birds. Proceedings of the Royal Society of London Series B 274: 561–568. 58 59 60 16 Biological Journal of the Linnean Society Page 17 of 40 Biological Journal of the Linnean Society
1 2 3 469 Immler S, Calhim S, Birkhead TR. 2008. Increased postcopulatory sexual selection reduces the intramale 4 470 variation in sperm design. Evolution 62: 1538–1543. 5 471 Immler S, Pitnick S, Parker GA, Durrant KL, Lüpold S, Calhim S, Birkhead TR. 2011. Resolving variation 6 472 in the reproductive tradeoff between sperm size and number .Proceedings of the National Academy of 473 SciencesUnited State of America 108: 5325–5330. 7 474 Immler S, Gonzalez Voyer A, Birkhead TR. 2012. Distinct evolutionary patterns of morphometric sperm traits 8 475 in passerine birds. Proceedings of the Royal Society of London Series B 279: 4174–4182. 9 476 Jamieson BGM, Ausio J, Justin J L. 1995. Advances in Spermatozoal Phylogeny and Taxonomy . 10 477 Paris:Museum National d'Histoire Naturelle. 11 478 Jamieson BGM. 2006. Avian spermatozoa: structure and phylogeny. In: Jamieson BGM, ed. Reproductive 12 479 biology and phylogeny of aves . Enfield, NH: Science Publishers Inc., 249–511. 13 480 Johansson US, Fjeldsa J, lokugalappatti LGS, Bowie RCK. 2007. A nuclear DNA phylogeny and proposed 14 481 taxonomic revision of African greenbuls (Aves, Passseriformes, Pycnonotidae). Zoologica Scripta 36: 15 482 417–427. 16 483 Jetz W, Thomas GH, Joy JB, Hartmann K, Mooers AO. 2012. The global diversity of birds in space and 17 484 time. Nature 491: 444–448. 18 485 Keith S, Urban EK, Fry CH. 1992. The birds of Africa . London: Academic Press. 486 Kleven O, LaskemoenFor T, Fossoy F, RobertsonPeer RJ, Lifj Revieweld JT. 2008. Intraspecific variation in sperm length is 19 487 negatively related to sperm competition in passerine birds. Evolution 62: 494–499. 20 488 Kleven O, Fossoy F, Laskemoen T, Robertson RJ, Rudolfsen G, Lifjeld JT. 2009. Comparative evidence for 21 489 the evolution of sperm swimming speed by sperm competition and female sperm storage duration in 22 490 passerine birds. Evolution 63: 2466–2473. 23 491 Koehler LD. 1995. Diversity of avian spermatozoa ultrastructure with emphasis on the members of ther order 24 492 Passeriformes. In Jamieson BGM, Ausio J, Justine J L, eds. Advances in Spermatozoal Phylogeny and 25 493 Taxonomy. Paris: Muséum National d’Histoire Naturelle, 437–444. 26 494 Laskemoen T, Kleven O, Fossøy F, Lifjeld JT. 2007. Intraspecific variation in sperm length in two passerine 27 495 species, the Bluethroat Luscinia svecica and the Willow Warbler Phylloscopus trochilus . Ornis Fennica 28 496 84: 131–139. 29 497 Laskemoen T, Kleven O, Fossøy F, Robertson RJ, Rudolfsen G, Lifjeld JT. 2010. Sperm quantity and 30 498 quality effects on fertilization success in a highly promiscuous passerine, the tree swallow Tachycineta 31 499 bicolor . Behavioral Ecology and Sociobiology 64(9): 1473–1483. Laskemoen T, Albrecht T, Bonisoli Alquati A, Cepak J, de Lope F, Hermosell I, Johannessen LE, Kleven 32 500 501 O, Marzal A, Mousseau TA, Møller AP, Robertson RJ, Rudolfsen G, Saino N, Vortman Y, Lifjeld 33 502 JT. 2013. Variation in sperm morphometry and sperm competition among barn swallow ( Hirundo 34 503 rustica ) populations. Behavioral Ecology and Sociobiology 67(2): 301–309. 35 504 Lifjeld JT, Laskemoen T, Kleven O, Albrecht T, Robertson RJ. 2010. Sperm length variation as a predictor 36 505 of extra pair paternity in passerine birds. PLoS ONE 5: e13456. 37 506 Macedo RH, Karubian J, Webster MS. 2008. Extra pair paternity and sexual selection in socially 38 507 monogamous birds: are tropical birds different? Auk 125: 769–777. 39 508 Morrow EH, Gage MJG. 2000. The evolution of sperm length in moths. Proceedings of the Royal Society of 40 509 London Series B 267: 307–313. 41 510 Moyle RG, Marks BD. 2006. Phylogenetic relationships of the bulbuls (Aves: Pycnonotidae) based on 42 511 mitochondrial and nuclear DNA sequence data. Molecular Phylogenetics and Evolution 40: 687–695. 43 512 Orme CDL, Freckleton RP, Thomas GH, Petzoldt T, Fritz SA, Isaac NJB, Pearse W. 2012. 44 513 caper:comparative analyses of phylogenetics andevolution in R. Methods in Ecology and Evolution 3: 45 514 145–151. 515 Parker GA. 1970. Sperm competition and its evolutionary consequenses in the insects. Biological Reviews 45: 46 516 525–567. 47 517 Pagel M. 1997. Infering evolutionary processes from phylogenies. Zoologica Scripta 26: 331–348. 48 518 Pagel M. 1999. Inferring the historical patterns of biological evolution. Nature 401: 877–884. 49 519 Pasquet E, Han L X, Khobket O, Cibois A. 2001. Towards a molecular systematics of the genus 50 520 Criniger , and a preliminary phylogeny of the bulbuls (Aves, Passeriformes, Pycnonotidae). 51 521 Zoosystema 23: 857–863. 52 522 Pitnick S, Hosken DJ, Birkhead TR. 2009. Sperm morphological diversity. In: Birkhead TR, 53 523 Hosken DJ, Pitnick S, eds. Sperm biology, an evolutionary perspective . San Diego, CA: 54 524 Academic Press, 69–149. 55 525 R Development Core Team. 2013. R: A language and environmen for statistical computing. R Foundation for 56 526 Statistical Computing . Available at: htt://www.R project.org. 57 527 Ratnasingham S, Hebert PDN. 2007. BOLD: The Barcode of Life Data System. Availaible at: 58 528 www.barcodinglife.org. Molecular Ecology Notes 7: 355–364. 59 60 17 Biological Journal of the Linnean Society Biological Journal of the Linnean Society Page 18 of 40
1 2 3 529 Rambaut A, Drummond AJ. 2009. TreeAnnotator. Version 1.5.4. 4 530 Reddy S. 2014. What’s missing from avian global diversification analyses? Molecular Phylogenetics and 5 531 Evolution 77: 159–165. 6 532 Revell LJ. 2012. Phytools: Phylogenetic Tools for Comparative Biology (and other things). Methods in Ecology 533 and Evolution 3: 217–223. 7 534 Revell LJ. 2013. Two new graphical methods for mapping trait evolution on phylogenies. Methods in Ecology 8 535 and Evolution 4: 754–759. 9 536 Rowe M, Albrecht T, Cramer ERA, Johnsen A, Laskemoen T, Weir JT, Lifjeld JT. 2015. Postcopulatory 10 537 sexual selection is associated with accelerated evolution of sperm morphology. Evolution 69: 1044– 11 538 1052. 12 539 Sibley CG, Ahlquist JE. 1990. Phylogeny and classification of birds. A study in molecular evolution. New 13 540 Haven & London: Yale University Press. 14 541 Snook RR. 2005. Sperm in competition: not playing by the numbers. Trends in Ecology and Evolution 20: 46– 15 542 53. 16 543 Sokal RR, Rohlf FJ. 1995. Biometry: the principles and practice of statistics in biological research . New York, 17 544 NY: Freeman. 18 545 Tamura K, Stecher ForG, Peterson D,Peer Filipski A, Kumar Review S. 2013. MEGA6: Molecular Evolutionary 19 546 Genetics Analysis version 6.0. Molecular Biology and Evolution 30: 2725–2729. 20 547 Tourmente M, Gomendio M, Roldan ERS. 2011. Sperm competition and the evolution of sperm design in 21 548 mammals. BMC Evolutionary Biology 11: 12. 549 Warton DI, Hui FKC. 2011. The arcsine is asinine: the analysis of proportions in ecology. Ecology 92: 3–10. 22 550 Westneat DF, Sherman PW. 1997. Density and extra pair fertilizations in birds: a comparative analysis. 23 551 Behavioral Ecology and Sociobiology 41: 205–215. 24 552 Wolfson A. 1952. The cloacal protuberance a means for determining breeding condition in live malepasserines. 25 553 Bird Banding 23: 159–165. 26 554 Zuccon D, Ericson PGP. 2010. The phylogenetic position of the Black collared Bulbul Neolestes torquatus . 27 555 Ibis 152: 386–392. 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48
49
50
51
52
53
54 55 56 57 58 59 60 18 Biological Journal of the Linnean Society Page 19 of 40 Biological Journal of the Linnean Society
1 2 3 4 SUPPORTING INFORMATION 5 6 Additional Supporting Information may be found in the online version of this article at the publisher's 7 web site: 8 9 Appendix S1. DNA extraction, COI gene sequencing and phylogeny construction of greenbuls. 10
11 12 Table S1. Detailed information of the 60 individuals of 21 species of greenbul used for the COI tree. 13 14 15 Table S2 Sperm morphology data for 167 individual male of 21 species of greenbul used in analysis. 16 17 Lengths (µm) of sperm head, midpiece, flagellum and total sperm are based on average of 10 18 spermatozoa measuredFor per individuals. Peer The CV wm isReview the coefficient of intra male variation of sperm 19 20 total length. 21 22 23 Table S3. Regression analysis controlling for phylogeny (PGLS) among sperm traits in 20 greenbuls 24 and one bulbul species. The model including the maximum likelihood of lambda ( λ) value was 25 26 compared against the models including λ = 1 and λ = 0, and superscripts following the λ values 27 indicate probability ( P) of likelihood ratio of sperm trait (first position: against λ = 0; second position: 28 29 against λ = 1). 30 31 32 Table S4. Regression analysis controlling for phylogeny (PGLS) between sperm traits and sperm 33 competition (sperm length CV ) among 10 greenbul and one bulbul species. The model including the 34 bm 35 maximum likelihood of lambda ( λ) value was compared against the models including λ = 1 and λ = 0, 36 and superscripts following the λ values indicate probability ( P) of likelihood ratio of sperm trait (first 37 38 position: against λ = 0; second position: against λ = 1). 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 19 Biological Journal of the Linnean Society Biological Journal of the Linnean Society Page 20 of 40
1 Table 1. Sperm morphology of 20 greenbuls and one bulbul species showing mean± standard deviation of sperm head, midpiece, flagellum and total length 2 3 (µm); included are intra male coefficient of variation (CV wm ) and inter male coefficient of variation of sperm length (CV bm ) and an ANOVA test of difference 4 5 between species 6 7 Species Country Head length Midpiece length Flagellum Total length CV wm (total CV bm (total 8 length length) length) 9 Phyllastrephus poensis (n = 3) Cameroon 12.50 ± 0.26 58.09 ± 2.08 67.29 ± 1.11 79.79 ± 0.98 2.06 ± 0.17 10 Phyllastrephus baumanni (n = 2) Nigeria 11.08 ± 0.44 51.22 ± 1.22 58.63 ± 1.19 69.71 ± 0.74 1.76 ± 0.55 11 Phyllastrephus albigularis (n = 17) Nigeria 11.88 ± 0.41 63.66 ± 2.33 73.31 ± 1.99 85.19 ± 2.13 1.98 ± 0.51 2.53 12 Phyllastrephus xavieri (n = 4) Cameroon 11.82 ± 0.47 66.63 ± 0.97 76.27 ± 3.12 88.09 ± 3.51 1.75 ± 0.42 4.23 13 Phyllastrephus icterinus (n = 5) Nigeria For 12.03 ±Peer 0.63 65.65 ± 2.47Review 76.67 ± 0.97 88.70 ± 1.36 2.04 ± 0.46 1.61 14 † 15 Phyllastrephus icterinus (n = 4) Cameroon 11.88 ± 0.38 64.36 ± 1.82 73.74 ± 1.73 85.62 ± 1.66 2.06 ± 0.36 2.06 16 Phyllastrephus poliocephalus (n = 1) Cameroon 12.41 60.64 68.62 81.03 1.77 17 Criniger calarus (n = 3) Cameroon 14.21 ± 0.75 36.75 ± 7.54 56.18 ± 2.30 70.39 ± 3.00 1.91 ± 0.91 18 †Criniger calarus (n = 1) Nigeria 15.33 44.64 56.95 72.28 1.90 19 Criniger barbatus (n = 1) Nigeria 14.51 78.59 87.71 102.22 1.96 20 Criniger chloronotus (n = 1) Cameroon 15.49 51.82 71.37 86.87 3.24 21 Eurillas ansorgei (n = 1) Cameroon 13.40 82..52 94.99 108.39 2.70 22 23 Eurillas gracilis (n = 1) Nigeria 14.65 91.28 101.80 116.45 2.27 24 Eurillas curvirostris (n = 7 ) Nigeria 15.64 ± 1.08 91.07 ± 1.05 101.53 ± 3.02 117.18 ± 3.67 1.75 ± 0.61 3.24 25 †Eurillas curvirostris (n = 2) Cameroon 15.27 90.64 98.25 113.52 1.40 26 Eurillas virens (n = 31) Nigeria 14.69 ± 0.70 79.38 ± 3.36 90.13 ± 3.48 104.81 ± 3.52 1.90 ± 0.64 3.38 27 †Eurillas virens (n = 1) Cameroon 15.40 75.18 87.99 103.40 2.32 28 Eurillas latirostris (n = 26) Nigeria 14.43 ± 0.68 85.90 ± 3.09 95.42 ± 3.47 109.85 ± 3.26 1.76 ± 0.64 3.00 29 Bleda syndactylus (n = 1) Nigeria 15.17 58.02 66.03 81.21 2.21 30 † 31 Bleda syndactylus (n = 1) Cameroon 14.95 59.57 71.14 86.09 3.08 32 Bleda canicapillus (n = 24) Nigeria 14.39 ± 0.58 62.47 ± 1.24 71.86 ± 2.32 86.26 ± 2.22 2.01 ± 0.62 2.61 33 Bleda notatus (n = 4 ) Cameroon 14.51 ± 1.16 63.61 ± 1.58 72.45 ± 2.30 86.97 ± 2.07 1.70 ± 0.42 2.38 34 Chlorocichla simplex (n = 2) Nigeria 14.07 ± 0.68 53.07 ± 0.15 64.71 ± 1.69 78.79 ± 2.37 2.26 ± 0.03 35 Arizelocichla montana (n = 7) Cameroon 13.73 ± 0.44 61.79 ± 2.49 72.82 ± 1.29 86.55 ± 1.68 2.02 ± 0.66 2.01 36 Arizelocichla tephrolaema (n = 5) Cameroon 13.54 ± 0.60 45.59 ± 1.46 56.90 ± 2.11 70.45 ± 2.41 3.27 ± 0.89 3.60 37 Pycnonotus barbatus (n = 20) Nigeria 13.47 ± 0.64 59.82 ± 2.01 70.21 ± 2.88 83.68 ± 3.11 2.37 ± 0.68 3.76 38 39 F20, 145 = 22.61 F20, 145 = 194.70 F20, 145 = 156.40 F20, 145 = 164.60 F20, 145 = 2.08 40 ANOVA P <0.0001 P <0.0001 P <0.0001 P <0.0001 P = 0.007 41 42 †Sperm morphology were not used in ANOVA and PGLS analysis 43 44 45 20 46 Biological Journal of the Linnean Society 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 21 of 40 Biological Journal of the Linnean Society
1 2 3 4 Table 2. Phylogenetic signal in sperm traits among 20 species of greenbul and one bulbul using 5 Pagel’s λ and Blomberg’s K with P values 6 7 8 Pagel’s λ Blomberg’s K 9 10 Sperm traits λ P (likelihood ration test) K P (randomization) 11 12 Head length 0.889 <0.001 0.464 0.105 13 Midpiece length 0.588 0.203 0.091 0.769 14 15 Flagellum length 0.804 0.017 0.203 0.441 16 17 Total length 0.833 0.013 0.246 0.362 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 21 Biological Journal of the Linnean Society Biological Journal of the Linnean Society Page 22 of 40
1 2 3 Table 3. AICc scores (AICc – AICc score for best fit model) and Akaike (AICc) weights showing 4 support for evolutionary models of sperm morphometrics in the Pycnonotidae. For each sperm trait, 5 6 the model with the lowest AICc value (i.e., AICc = 0) is considered the best fitting model (boldface 7 2 8 with *). The parameters estimated by the models are: σ = net rate of trait evolution in Brownian 9 motion model or the initial rate of evolution in the Early Burst model, λ = extent to which phylogeny 10 11 predicts covariance among trait for species, δ = compares contribution of early versus late trait 12 evolution across a phylogeny, κ = evolutionary change in trait associated with speciation events along 13 14 the branch length in the Kappa models, α = evolutionary constraint parameter in the Ornstein 15 16 Uhlenbeck model moving trait values back to the optimum and r = change in rate of trait evolution 17 through time in the Early Burst model. See details of model parameters in the methods 18 For Peer Review 19 Length of sperm traits 20 21 Evolutionary Parameters Head length Midpiece Flagellum Total 22 models length length length 23 Brownian AICc 21.66 39.43 29.23 25.27 24 motion AICc weight <0.0001 <0.0001 <0.0001 <0.0001 (BM) 2 25 σ 0.0009 0.0280 0.0075 0.0048 26
27 28 Lambda AICc 6.45 0.00* 2.32 2.01 29 AICc weight 0.0365 0.4213 0.2357 0.2644 30 λ 0.89 0.59 0.80 0.83 31 2 σ 0.0003 0.0021 0.0012 0.0010 32 33 34 Delta AICc 18.08 30.15 21.92 18.68 35 AICc weight 0.0001 <0.0001 <0.0001 <0.0001 36 δ 2.99 2.99 2.99 2.99 37 σ2 0.0004 0.0102 0.0029 0.0020 38 39 40 Kappa AICc 0.00* 0.15 0.00* 0.00* 41 AICc weight 0.9631 0.3913 0.7503 0.7231 42 κ 0.00 <0.0001 <0.0001 <0.0001 43 σ2 0.0015 0.0199 0.0086 0.0067 44 45 46 Ornstein AICc 16.35 1.62 7.97 8.13 47 Uhlenbeck AICc weight 0.0003 0.1874 0.0139 0.0124 (OU) 48 α 0.108 62.621 55.947 55.947 49 σ2 0.0015 6.3912 3.3466 2.6561 50 51 52 Early Burst AICc 24.45 42.22 32.03 28.06 53 (EB) AICc weight <0.0001 <0.0001 <0.0001 <0.0001 54 r 0.00 0.00 0.00 0.00 55 σ2 56 0.0009 0.0280 0.0075 0.0049 57 58 59 60 22 Biological Journal of the Linnean Society Page 23 of 40 Biological Journal of the Linnean Society
1 2 3 FIGURE LEGENDS 4 5 6 7 Figure 1. Relationship between sperm total length and sperm head, midpiece and flagellum length 8 9 among greenbuls including one bulbul (N = 21 species). Each data point represents the species mean 10 11 for each sperm trait. 12 13 14 15 Figure 2. Ancestral character estimation and variation in (A) sperm total length and (B) sperm head 16 17 length along the branches and nodes of the phylogeny of 20 study species of greenbul and one bulbul. 18 For Peer Review 19 Number on the scale bars represents the range of sperm total length and sperm head length 20 21 respectively for the species. The scale bar for colours also indicates the scale for branch lengths in 22 23 million years (Myr). 24 25 26 27 Figure 3. Traitgram showing the projection of the greenbul phylogeny into a space defined by sperm 28 29 total length (µm) (y axis) and node age, i.e. time since divergence from the root (x axis). The vertical 30 31 position of nodes and branches are computed via ancestral character estimation using maximum 32 33 likelihood. Embedded figure showed uncertainty through increasing transparency of the plotted blue 34 35 lines around the point estimates with the entire range showing the 95% confidence interval. 36 37
38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 23 Biological Journal of the Linnean Society Biological Journal of the Linnean Society Page 24 of 40
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 Figure 1. Relationship between sperm total length and sperm head, midpiece and flagellum length among 37 greenbuls including one bulbul (N = 21 species). Each data point represents the species mean for each 38 sperm trait. 112x96mm (300 x 300 DPI) 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Biological Journal of the Linnean Society Page 25 of 40 Biological Journal of the Linnean Society
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 Figure 2. Ancestral character estimation and variation in (A) sperm total length and (B) sperm head length 26 along the branches and nodes of the phylogeny of 20 study species of greenbul and one bulbul. Number on 27 the scale bars represents the range of sperm total length and sperm head length respectively for the 28 species. The scale bar for colours also indicates the scale for branch lengths in million years (Myr). 29 1325x702mm (96 x 96 DPI) 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Biological Journal of the Linnean Society Biological Journal of the Linnean Society Page 26 of 40
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 Figure 3. Traitgram showing the projection of the greenbul phylogeny into a space defined by sperm total 35 length (µm) (y axis) and node age, i.e. time since divergence from the root (x axis). The vertical position of 36 nodes and branches are computed via ancestral character estimation using maximum likelihood. Embedded 37 figure showed uncertainty through increasing transparency of the plotted blue lines around the point 38 estimates with the entire range showing the 95% confidence interval. 749x605mm (96 x 96 DPI) 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Biological Journal of the Linnean Society Page 27 of 40 Biological Journal of the Linnean Society
1 2 3 SUPPORTING INFORMATION 4 5 6 Sperm size evolution in African greenbuls (Passeriformes: Pycnonotidae) 7 8 9 TAIWO C. OMOTORIOGUN, TOMAS ALBRECHT, DAVID HOŘÁK, TERJE LASKEMOEN, 10 11 ULF OTTOSSON, MELISSAH ROWE, ONDŘEJ SEDLÁČEK AND JAN T. LIFJELD 12 13 Appendix S1. DNA extraction, COI gene sequencing and phylogeny construction of greenbuls 14 15 We sequenced mtDNA from the first part of the COI gene, between 650 and 750 bp in length for 18 16 17 species of greenbul. DNA was extracted from blood following the protocol for the E.Z.N.A blood kit 18 (Omega Bio Tek, Inc.,For Norcross, Georgia).Peer The PCR Review reaction set ( c. 12.5 µl) contained dH 2O, 1X PCR 19 20 buffer (20 mM Tris HCl, 50 mM KCl; Invitrogen), 2.5 mM MgCl 2, 0.5 mM dNTP, 0.1 µM forward 21 and reverse primers, 0.3 U Platinum Taq polymerase (Invitrogen) and DNA sample. Primers were 22 23 either PasserF1 (5’ CCAACCACAAAGACATCGGAACC 3’) and PasserR1 (5’ 24 GTAAACTTCTGGGTGACCAAAGAATC 3’) (Lohman et al. 2008), and/or a combination of CO1 25 26 ExtF (5’ ACGCTTTAACACTCAGCCATCTTACC 3’) (Johnsen et al. 2010) or BirdF1 (5’ 27 TTCTCCAACCACAAAGACATTGGCAC 3’) (Hebert et al. 2004) and COIbirdR2 (5’ 28 29 ACGTGGGAGATAATTCCAAATCCTGG 3’) (Hebert et al. 2004) or BirdR1 (5’ 30 31 ACGTGGGAGATAATTCCAAATCCTG 3’) (Hebert et al. 2004) in the case where primers did not 32 work well. Primers were amplified in a thermocycler using the following profile settings: 94 o C for 2 33 o o o o 34 min, 35 cycles of 94 C for 30 s, 55 C for 30 s and 72 C for 45 s, and in the end 72 C for 7 min. 35 PCR products were electrophoresed in 1% agarose TBE to quantify amplification success and exclude 36 37 any contamination. We then digested unexpended nucleotides and primers in the remaining PCR 38 product using Exosap IT (1:10, of Exonuclease I and Shrimp Alkaline Phosphatase) and allowed it to 39 o o 40 run at 37 C for 45 min and at 80 C for 15 min to deactivate the enzyme. The sequencing reactions 41 were performed by StarSEQ GmbH (Germany) and the sequences were proofread in CodonCode 42 43 Aligner v3.7.1 (CodonCode Corporation). Sequences and supporting information are available in the 44 BOLD database (Ratnasingham and Hebert 2007) in the project folder NHM Birds Greenbul 45 46 (BONGR) .We aligned mtDNA sequences using ClustalW in the program MEGA version 6.06 47 (Tamura et al. 2013). We applied Kimura 2 parameter model and 10000 bootstrap iterations to 48 49 construct a maximum likelihood tree with branch lengths (Fig S1). 50 51 52 53 54 55 56 57 58 59 60 Biological Journal of the Linnean Society Biological Journal of the Linnean Society Page 28 of 40
1 2 3 4 5 6 7
8
9
10
11
12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 Figure S1. Maximum Likelihood (Kimura 2 parameter model, 10000 bootstrap iterations) tree 52 53 topology based on mitochondria (COI) sequences from 61 individual of 18 study species of greenbul. 54 55 56 57 58 59 60 Biological Journal of the Linnean Society Page 29 of 40 Biological Journal of the Linnean Society
1 2 3 4 5 Table S1. Detail information of the 60 individuals of 21 species of greenbul used for the COI tree 6 7 Species Collection date Country Locality Accession no* BOLD ID 8 Eurillas curvirostris 09 Jun 2012 Nigeria Ibadan NHMO BI 32471 BONAF610 13 9 Eurillas curvirostris 21 Apr 2011 Nigeria Ibadan NHMO BI 29279 BONAF384 12 10 Eurillas curvirostris 26 Apr 2011 Nigeria Ibadan NHMO BI 29322 BONAF397 12 11 Eurillas curvirostris 20 Apr 2011 Nigeria Ibadan NHMO BI 29273 BONAF380 12 12 Eurillas curvirostris 29 May 2012 Nigeria Ogun NHMO BI 32323 BONAF584 13 13 Stelgidillas gracilirostris 16 Nov 2010For Cameroon Peer Laide farm ReviewNHMO BI 26079 BONAF185 12 14 Stelgidillas gracilirostris 23 Nov 2010 Cameroon Laide farm NHMO BI 26321 BONAF261 12 15 Stelgidillas gracilirostris 23 Nov 2010 Cameroon Laide farm NHMO BI 26322 BONAF262 12 16 Stelgidillas gracilirostris 23 Nov 2010 Cameroon Laide farm NHMO BI 26325 BONAF263 12 17 Eurillas gracilis 24 May 2012 Nigeria Ogun NHMO BI 32207 BONAF565 13 18 Eurillas latirostris 16 Aug 2012 Nigeria Benin NHMO BI 33740 BONAF711 13 19 Eurillas latirostris 15 Apr 2011 Nigeria Ibadan NHMO BI 29161 BONAF358 12 20 Eurillas latirostris 16 Aug 2012 Nigeria Benin NHMO BI 33737 BONAF708 13 21 Eurillas latirostris 15 Apr 2011 Nigeria Ibadan NHMO BI 29160 BONAF357 12 22 Eurillas latirostris 16 Aug 2012 Nigeria Benin NHMO BI 33741 BONAF712 13 23 Arizelocichla montanus 23 Nov 2010 Cameroon Laide farm NHMO BI 26320 BONAF260 12 24 Arizelocichla montanus 23 Nov 2010 Cameroon Laide farm NHMO BI 26317 BONAF258 12 25 Arizelocichla montanus 24 Nov 2010 Cameroon Laide farm NHMO BI 26364 BONAF273 12 26 Arizelocichla montanus 23 Nov 2010 Cameroon Laide farm NHMO BI 26316 BONAF257 12 27 Arizelocichla tephrolaemus 22 Nov 2010 Cameroon Laide farm NHMO BI 26308 BONAF253 12 28 Arizelocichla tephrolaemus 24 Nov 2008 Cameroon Oku NHMO BI 18829 BONAF085 12 29 Arizelocichla tephrolaemus 22 Nov 2008 Cameroon Oku NHMO BI 18819 BONAF082 12 30 Arizelocichla tephrolaemus 20 Nov 2008 Cameroon Oku NHMO BI 18813 BONAF078 12 31 Eurillas virens 18 Apr 2011 Nigeria Ibadan NHMO BI 29248 BONAF374 12 32 Eurillas virens 18 May 2012 Nigeria Ogun NHMO BI 32061 BONAF546 13 33 Eurillas virens 18 Apr 2011 Nigeria Ibadan NHMO BI 29226 BONAF370 12 34 Eurillas virens 16 Aug 2012 Nigeria Benin NHMO BI 33738 BONAF709 13 35 Bleda canicapillus 16 Apr 2011 Nigeria Ibadan NHMO BI 29186 BONAF363 12 36 Bleda canicapillus 20 Apr 2011 Nigeria Ibadan NHMO BI 29276 BONAF382 12 37 Bleda canicapillus 15 Apr 2011 Nigeria Ibadan NHMO BI 29177 BONAF361 12 38 39 40 41 42 43 44 45 46 Biological Journal of the Linnean Society 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Biological Journal of the Linnean Society Page 30 of 40
1 2 3 4 5 Table S1. Continued 6 7 Species Collection date Country Lo cality Accession no* BOLD ID 8 Bleda canicapillus 16 Aug 2012 Nigeria Benin NHMO BI 33733 BONAF704 13 9 Bleda syndactylus 29 May 2012 Nigeria Ogun NHMO BI 32329 BONAF588 13 10 Bleda syndactylus 15 Aug 2012 Nigeria Benin NHMO BI 33712 BONAF703 13 11 Atimastillas flavicollis 08 Mar 2011 Nigeria Jos NHMO BI 29035 BONAF333 12 12 Atimastillas flavicollis 21 Jan 2011 Nigeria Jos NHMO BI 28819 BONAF299 12 13 Atimastillas flavicollis 08 Mar 2011For Nigeria Peer Jos ReviewNHMO BI 29043 BONAF334 12 14 Chlorocichla simplex 24 May 2011 Nigeria Ibadan NHMO BI 29376 BONAF409 12 15 Chlorocichla simplex 06 Jun 2012 Nigeria Ibadan NHMO BI 32396 BONAF604 13 16 Chlorocichla simplex 23 May 2012 Nigeria Ogun NHMO BI 32192 BONAF564 13 17 Chlorocichla simplex 22 May 2012 Nigeria Ogun NHMO BI 32171 BONAF563 13 18 Criniger barbatus 22 May 2012 Nigeria Ogun NHMO BI 32136 BONAF556 13 19 Criniger calurus 05 Jul 2013 Nigeria Ogun NHMO BI 36156 BONAF783 13 20 Criniger calurus 19 May 2012 Nigeria Ogun NHMO BI 32111 BONAF553 13 21 Criniger calurus 19 May 2012 Nigeria Ogun NHMO BI 32109 BONAF552 13 22 Phyllastrephus albigularis 21 Apr 2011 Nigeria Ibadan NHMO BI 29287 BONAF505 12 23 Phyllastrephus albigularis 16 Apr 2011 Nigeria Ibadan NHMO BI 29187 BONAF364 12 24 Phyllastrephus albigularis 20 May 2011 Nigeria Ibadan NHMO BI 29360 BONAF507 12 25 Phyllastrephus baumanni 19 Apr 2011 Nigeria Ibadan NHMO BI 29266 BONAF377 12 26 Phyllastrephus baumanni 06 Jun 2012 Nigeria Ibadan NHMO BI 32407 BONAF607 13 27 Phyllastrephus baumanni 06 Jun 2012 Nigeria Ibadan NHMO BI 32406 BONAF606 13 28 Phyllastrephus baumanni 20 May 2011 Nigeria Ibadan NHMO BI 29362 BONAF534 13 29 Phyllastrephus icterinus 27 May 2012 Nigeria Ogun NHMO BI 32270 BONAF579 13 30 Phyllastrephus icterinus 27 May 2012 Nigeria Ogun NHMO BI 32269 BONAF578 13 31 Phyllastrephus icterinus 27 May 2012 Nigeria Ogun NHMO BI 32263 BONAF577 13 32 Phyllastrephus icterinus 27 May 2012 Nigeria Ogun NHMO BI 32252 BONAF574 13 33 Phyllastrephus scandens 23 Apr 2011 Nigeria Ibadan NHMO BI 29315 BONAF395 12 34 Pycnonotus barbatus 28 Apr 2010 Nigeria Jos NHMO BI 25693 BONAF089 12 35 Pycnonotus barbatus 15 Nov 2008 Cameroon Big Babanki NHMO BI 18784 BONAF072 12 36 Pycnonotus barbatus 10 Nov 2008 Cameroon Big Czech Babanki NHMO BI 18751 BONAF068 12 37 Pycnonotus barbatus 05 Nov 2008 Cameroon Big Czech Babanki NHMO BI 18727 BONAF064 12 38 39 *Bird collection at the Natural History Museum, Oslo 40 41 42 43 44 45 46 Biological Journal of the Linnean Society 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 31 of 40 Biological Journal of the Linnean Society
1 2 3 4 5 Table S2. Sperm morphology data for 167 individual male of 21 species of greenbul used in analysis. Length (µm) of sperm head, midpiece, flagellum and total 6 7 sperm total are based on average of 10 spermatozoa measured per individuals. The CV wm is coefficient of intra male variation of sperm total length 8 Sperm head Sperm midpiece Sperm flagellum Sperm total 9 Species Accession no Collection date Country length length length length CV wm 10 Eurillas curvirostris NHMO BI 29273 20/04/2011 Nigeria 13.75 93.033 99.235 112.985 1.923 11 12 Eurillas curvirostris NHMO BI 29279 21/04/2011 Nigeria 15.118 91.068 99.091 114.209 2.548 13 Eurillas curvirostris NHMO BI 29322For 26/04/2011 Peer Nigeria Review 17.297 91.012 102.877 120.173 1.296 14 Eurillas curvirostris NHMO BI 29340 18/05/2011 Nigeria 15.547 91.697 104.52 120.067 1.427 15 Eurillas curvirostris NHMO BI 29363 20/05/2011 Nigeria 15.656 90.577 97.008 112.664 2.588 16 Eurillas curvirostris NHMO BI 32323 29/05/2012 Nigeria 16.067 90.05 104. 001 120.069 1.223 17 18 Eurillas curvirostris NHMO BI 32323 29/05/2012 Nigeria 16.067 90.049 104.001 120.068 1.223 19 † Eurillas curvirostris NHMO BI 44594 19/12/2012 Cameroon 15.794 87.93 96.05 111.844 1.604 20 † Eurillas curvirostris NHMO BI 83580 01/11/2013 Cameroon 14.751 93.368 100.450 115.202 1.211 21 Eurillas gracilis NHMO BI 36134 03/07/2013 Nigeria 14.645 91.276 101.803 116.448 2.267 22 Eurillas latirostris NHMO BI 43618 19/04/2011 Nigeria 13.783 87.933 95.864 109.647 2.052 23 24 Eurillas latirostris NHMO BI 32115 19/05/2012 Nigeria 14.143 85.24 93.45 107.593 1.92 0 25 Eurillas latirostris NHMO BI 32142 22/05/2012 Nigeria 14.33 82.497 94.415 108.746 2.535 26 Eurillas latirostris NHMO BI 32462 09/06/2012 Nigeria 14.144 86.424 94.718 108.861 2.061 27 Eurillas latirostris NHMO BI 32467 09/06/2012 Nigeria 15.489 78.76 85.946 101.435 1.445 28 09/06/2012 29 Eurillas latirostris NHMO BI 32475 Nigeria 14.496 87.73 95.715 110.21 1.323 30 Eurillas latirostris NHMO BI 32478 11/06/2012 Nigeria 15.216 82.033 91.35 106.566 1.330 31 Eurillas latiro stris NHMO BI 32486 11/06/2012 Nigeria 13.229 86.72 93.908 107.136 1.599 32 Eurillas latirostris NHMO BI 32513 12/06/2012 Nigeria 15.778 86.58 96.29 112.068 2.095 33 Eurillas latirostris NHMO BI 32515 12/06/2012 Nigeria 14.154 82.431 89.913 104.067 2.331 34 12/06/2012 35 Eurillas latirostris NHMO BI 32519 Nigeria 14.254 84.674 94.814 109.068 1.543 36 Eurillas latirostris NHMO BI 32539 12/06/2012 Nigeria 13.968 85.53 97.662 111.63 2.019 37 Eurillas latirostris NHMO BI 32541 12/06/2012 Nigeria 14.511 79.813 97.539 112.05 1.233 38 Eurillas latirostris NHMO BI 32568 15/06/2012 Nigeria 13.826 86.34 97.463 111.288 0.713 39 Eurillas latirostris NHMO BI 32572 15/06/2012 Nigeria 13.801 86.686 93.744 107.544 1.608 40 41 Eurillas latirostris NHMO BI 33634 11/08/2012 Nigeria 14.693 85.582 96 .057 110.749 1.303 42 43 44 45 46 Biological Journal of the Linnean Society 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Biological Journal of the Linnean Society Page 32 of 40
1 2 3 4 5 6 7 Table S2. Continued 8 Sperm head Sperm midpiece Sperm flagellum Sperm total 9 Species Accession no Collection date Country length length length length CV wm 10 Eurillas latirostris NHMO BI 33646 11/08/2012 Nigeria 14.511 85.951 94.255 108.766 1.566 11 12 Eurillas latirostris NHMO BI 33713 15/08/2012 Nigeria 13.902 88.24 103.368 117.27 1.818 13 Eurillas latirostris NHMO BI 33745For 16/08/2012 Peer Nigeria Review 15.709 85.408 93.098 108.808 2.338 14 Eurillas latirostris NHMO BI 36125 03/07/2013 Nigeria 13.713 92.298 99.068 112.78 1.198 15 Eurillas latirostris NHMO BI 36055 26/06/2013 Nigeria 14.838 85.134 93.716 108.554 1.41 16 Eurillas latirostris NHMO BI 36057 26/06/2013 Nigeria 14.379 88.505 95.502 109.881 1.551 17 18 Eurillas latirostris NHMO BI 36126 03/07/2013 Nigeria 14.458 90.581 100.155 114.613 1.038 19 Eurillas latirostris NHMO BI 36127 02/07/2013 Nigeria 15.168 84.493 95.091 110.259 4.008 20 Eurillas latirostris NHMO BI 36128 03/07/2013 Nigeria 13.44 90.356 100.691 114.131 1.494 21 Eurillas latirostris NHMO BI 36135 04/07/2013 Nigeria 15.229 87.532 97.022 112.25 2.228 22 Eurillas virens NHMO BI 29226 18/04/2011 Nigeria 13.901 79.063 87.441 101.342 1.089 23 24 Eurillas virens NHMO BI 29248 18/04/2011 Nigeria 14.752 75.332 83.522 98.274 3.95 0 25 Eurillas virens NHMO BI 43621 22/04/2011 Nigeria 14.967 82.988 94.346 109.313 1.438 26 Eurillas virens NHMO BI 43622 25/04/2011 Nigeria 15.16 82.584 91.069 106.229 2.093 27 Eurillas virens NHMO BI 43625 26/04/2011 Nigeria 13.4 77.834 92.854 106.254 3.174 28 24/05/2011 29 Eurillas virens NHMO BI 43637 Nigeria 14.397 74.127 82.978 97.375 1.624 30 Eurillas virens NHMO BI 29377 24/05/2011 Nigeria 14.1 75.361 87.926 102.026 1.393 31 Eurillas virens NHMO BI 29389 26/05/2011 Nigeria 15.355 78.833 89.955 105.31 1.752 32 Eurillas virens NHMO BI 29397 27/05/2011 Nigeria 15.376 77.368 84.44 99.816 2.161 33 Eurillas virens NHMO BI 32073 18/05/2012 Nigeria 16.378 82.135 94.814 111.192 2.101 34 18/05/2012 35 Eurillas virens NHMO BI 32087 Nigeria 14.297 76.397 90.441 104.738 1.303 36 Eurillas virens NHMO BI 32103 19/05/2012 Nigeria 15.187 80.646 88.714 103.9 1.727 37 Eurillas virens NHMO BI 32120 19/05/2012 Nigeria 14.806 81.798 89.547 104.353 1.772 38 Eurillas virens NHMO BI 32128 19/05/2012 Nigeria 15.396 82.098 93.771 109.167 2.495 39 Eurillas virens NHMO BI 32141 22/05/2012 Nigeria 14.46 80.43 89.972 104.432 1.247 40
41
42 43 44 45 46 Biological Journal of the Linnean Society 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 33 of 40 Biological Journal of the Linnean Society
1 2 3 4 5 Table S2. Continued 6 Sperm head Sperm midpiece Sperm flagellum Sperm total 7 Species Accession no Collection date Country length length length length CV wm 8 Eurillas virens NHMO BI 32183 23/05/2012 Nigeria 14.94 77.184 91.716 106.656 1.482 9 23/05/2012 10 Eurillas virens NHMO BI 32200 Nigeria 13.988 87.887 97.745 111.734 1.46 11 Eurillas virens NHMO BI 32206 24/05/2012 Nigeria 14.406 78.281 86.98 101.386 1.791 12 Eurillas virens NHMO BI 32205 24/05/2012 Nigeria 14.538 80.958 89.392 103.93 2.601 13 Eurillas virens NHMO BI 32240For 25/05/2012 Peer Nigeria Review 15.672 77.818 91.313 106.985 1.567 14 Eurillas virens NHMO BI 32217 24/05/2012 Nigeria 13.805 75.198 86.771 100.576 1.386 15 16 Eurillas virens NHMO BI 32235 24/05/2012 Nigeria 13.716 80.453 92.235 105.951 1.976 17 Eurillas virens NHMO BI 32255 27/05/2012 Nigeria 15.607 71.496 86.888 102.494 2.223 18 Eurillas virens NHMO BI 32399 06/06/2012 Nigeria 14.892 76.622 88.771 103.663 2.200 19 Eurillas virens NHMO BI 32506 10/06/2012 Nigeria 15.168 82.396 88.631 103.798 1.972 20 Eurillas virens NHMO BI 32508 10/06/2012 Nigeria 14.158 82.781 96.982 111.14 1.884 21 22 Eurillas virens NHMO BI 32544 12/06/2012 Nigeria 14.959 78.577 89.942 104.9 3.024 23 Eurillas virens NHMO BI 32551 14/06/2012 Nigeria 15.201 80.637 89.609 104.809 0.988 24 Eurillas virens NHMO BI 35994 21/06/2013 Nigeria 13.488 78.39 90.769 104.257 1.731 25 Eurillas virens NHMO BI 35981 20/06/2013 Nigeria 14.201 82.83 92.738 106.939 1.468 26 Eurillas virens NHMO BI 36001 21/06/2013 Nigeria 14.612 82.247 91.667 106.279 1.798 27 † 28 Eurillas virens NHMO BI 46085 07/03/2012 Cameroon 15.399 75.167 87.997 103.396 2.315 29 Bleda canicapillus NHMO BI 29177 15/04/2011 Nigeria 14.148 61.984 73.411 87.559 1.421 30 Bleda canicapillus NHMO BI 29186 16/04/2011 Nigeria 13.463 61.04 72.113 85.577 3.852 31 Bleda canicapillus NHMO BI 29276 20/04/2011 Nigeria 14.098 62.662 69.894 83.992 2.127 32 18/05/2011 33 Bleda canicapillus NHMO BI 43627 Nigeria 14.87 60.832 73.986 88.856 2.776 34 Bleda canicapillus NHMO BI 43631 18/05/2011 Nigeria 15.008 63.349 72.134 87.142 1.158 35 Bleda canicapillus NHMO BI 43638 23/05/2011 Nigeria 13.984 62.367 72.244 86.228 2.775 36 Bleda canicapillus NHMO BI 32108 19/05/2012 Nigeria 14.638 62.507 72.376 87.014 2.188 37 Bleda canicapillus NHMO BI 32288 28/05/2012 Nigeria 14.292 63.83 71.92 86.212 1.589 38 09/06/2012 39 Bleda canicapillus NHMO BI 32472 Nigeria 14.739 60.721 68.375 83.114 1.788 40 41 42 43 44 45 46 Biological Journal of the Linnean Society 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Biological Journal of the Linnean Society Page 34 of 40
1 2 3 4 5 Table S2. Continued 6 Sperm head Sperm midpiece Sperm flagellum Sperm total 7 Species Accession no Collection date Country length length length length CV wm 8 Bleda canicapillus NHMO BI 32518 12/06/2012 Nigeria 15.193 62.618 69.277 84.47 1.616 9 14/06/2012 10 Bleda canicapillus NHMO BI 32550 Nigeria 14.688 61.824 67.675 82.362 1.487 11 Bleda canicapillus NHMO BI 32556 14/06/2012 Nigeria 14.109 63.053 69.429 83.538 3.013 12 Bleda canicapillus NHMO BI 33611 09/08/2012 Nigeria 15.053 59.596 69.453 84.506 1.626 13 Bleda canicapillus NHMO BI 33624For 10/08/2012 Peer Nigeria Review 13.832 62.653 74.404 88.236 1.496 14 Bleda canicapillus NHMO BI 33693 14/08/2012 Nigeria 15.165 63.347 71.005 86.169 2.005 15 16 Bleda canicapillus NHMO BI 35992 21/06/2013 Nigeria 14.081 60.458 69.564 83.645 2.010 17 Bleda canicapillus NHMO BI 36027 26/06/2013 Nigeria 14.716 63.08 74.481 89.196 1.871 18 Bleda canicapillus NHMO BI 36012 24/06/2013 Nigeria 13.812 62.759 73.243 87.055 1.838 19 Bleda canicapillus NHMO BI 35999 21/06/2013 Nigeria 14.126 62.943 71.7 85.826 1.38 20 Bleda canicapillus NHMO BI 35959 19/06/2013 Nigeria 13.063 62.614 72.856 85.919 2.077 21 22 Bleda canicapillus NHMO BI 35973 18/06/2013 Nigeria 14.777 62.349 70.043 84.82 1.878 23 Bleda canicapillus NHMO BI 36006 22/06/2013 Nigeria 15.294 63.998 72.96 88.254 1.651 24 Bleda canicapillus NHMO BI 35905 16/06/2013 Nigeria 14.027 64.927 75.228 89.256 1.853 25 Bleda canicapillus NHMO BI 36005 22/06/2013 Nigeria 14.251 63.694 76.958 91.208 2.703 26 Bleda syndactylus NHMO BI 33654 11/08/2012 Nigeria 15.175 58.02 66.033 81.207 2.209 27 † 28 Bleda syndactylus NHMO BI 83579 01/11/2013 Cameroon 14.946 59.569 71.140 86.086 3.079 29 Bleda notatus NHMO BI 83519 01/11/2013 Cameroon 15.012 65.410 74.039 89.052 1.567 30 Bleda notatus NHMO BI 83520 01/11/2013 Cameroon 14.445 62.295 69.666 84.111 2.246 31 Bleda notatus NHMO BI 83521 01/11/2013 Cameroon 15.672 64.463 71.472 87.144 1.722 32 33 Bleda notatus NHMO BI 46082 07/03/2012 Cameroon 12.940 62.264 74.617 87.557 1.252 34 Chlorocichla simplex NHMO BI 36088 02/07/2013 Nigeria 13.591 53.176 63.516 77.107 2.243 35 Chlorocichla simplex NHMO BI 36188 07/07/2013 Nigeria 14.557 52.96 65.908 80.465 2.279 36 Criniger barbatus NHMO BI 32175 23/05/2012 Nigeria 14.511 78.591 87.709 102.22 1.957 37 †Criniger calurus NHMO BI 32267 27/05/2012 Nigeria 15.329 44.636 56.953 72.282 1.901 38 01/11/2013 39 Criniger calurus NHMO BI 83531 Cameroon 15.056 32.521 58.794 73.850 1.477 40 Criniger calurus NHMO BI 83532 01/11/2013 Cameroon 13.606 32.273 55.229 68.836 1.306 41 Criniger calurus NHMO BI 83533 01/11/2013 Cameroon 13.973 45.463 54.504 68.477 2.953 42 43 44 45 46 Biological Journal of the Linnean Society 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 35 of 40 Biological Journal of the Linnean Society
1 2 3 4 5 Table S2. Continued 6 Sperm head Sperm midpiece Sperm flagellum Sperm total 7 Species Accession no Collection date Country length length length length CV wm 8 Phyllastrephus albigularis NHMO BI 29187 16/04/2011 Nigeria 11.426 65.164 74.678 86.104 1.091 9 19/04/2011 10 Phyllastrephus albigularis NHMO BI 29261 Nigeria 11.882 62.45 74.827 86.709 1.482 11 Phyllastrephus albigularis NHMO BI 29187 16/04/2011 Nigeria 12.059 64.901 74.889 86.948 1.564 12 Phyllastrephus albigularis NHMO BI 29287 21/04/2011 Nigeria 11.302 65.927 75.906 87.208 2.383 13 Phyllastrephus albigularis NHMO BI 29360For 20/05/2011Peer Nigeria Review 12.845 57.34 73.836 86.681 2.712 14 Phyllastrephus albigularis NHMO BI 43639 23/05/2011 Nigeria 12.046 63.405 75.141 87.187 2.121 15 16 Phyllastrephus albigularis NHMO BI 32086 18/05/2012 Nigeria 11.524 65.797 72.075 83.599 2.511 17 Phyllastrephus albigularis NHMO BI 32294 28/05/2012 Nigeria 11.166 65.478 70.903 82.068 2.359 18 Phyllastrephus albigularis NHMO BI 32339 30/05/2012 Nigeria 11.737 64.311 71.517 83.254 1.460 19 Phyllastrephus albigularis NHMO BI 32453 09/06/2012 Nigeria 12.251 64.798 72.836 85.087 1.837 20 Phyllastrephus albigularis NHMO BI 32480 11/06/2012 Nigeria 11.92 62.784 69.759 81.679 3.009 21 22 Phyllastrephus albigularis NHMO BI 33601 09/08/2012 Nigeria 11.711 59.739 69.724 81.436 1.423 23 Phyllastrephus albigularis NHMO BI 33660 12/08/2012 Nigeria 11.831 62.464 71.579 83.411 1.803 24 Phyllastrephus albigularis NHMO BI 36016 24/06/2013 Nigeria 11.907 62.318 74.498 86.405 1.968 25 Phyllastrephus albigularis NHMO BI 46241 21/06/2013 Nigeria 11.91 65.72 75.006 86.916 1.925 26 Phyllastrephus albigularis NHMO BI 35951 19/06/2013 Nigeria 11.963 64.185 74.258 86.221 1.789 27 28 Phyllastrephus albigularis NHMO BI 35907 16/06/2013 Nigeria 12.427 65.411 74.903 87.33 2.216 29 Phyllastrephus baumanni NHMO BI 29266 19/04/2011 Nigeria 10.766 52.081 59.477 70.243 1.371 30 Phyllastrephus baumanni NHMO BI 29379 24/05/2011 Nigeria 11.395 50.362 57.789 69.184 2.149 31 Phyllastrephus icterinus NHMO BI 32273 27/05/2012 Nigeria 12.407 68.124 77.576 89.983 1.899 32 29/05/2012 33 Phyllastrephus icterinus NHMO BI 32333 Nigeria 12.087 68.115 77.411 89.498 2.251 34 Phyllastrephus icterinus NHMO BI 32333 29/05/2012 Nigeria 12.443 65.42 77.068 89.511 1.755 35 Phyllastrephus icterinus NHMO BI 32352 30/05/2012 Nigeria 12.301 64.029 75.389 87.69 1.567 36 Phyllastrephus icterinus NHMO BI 33656 12/08/2012 Nigeria 10.933 62.564 75.903 86.837 2.727 37 † Phyllastrephus icterinus NHMO BI 44595 29/12/2011 Cameroon 11.44 65.829 74.03 85.47 2.552 38 † 28/12/2011 39 Phyllastrephus icterinus NHMO BI 44617 Cameroon 11.943 63.452 76.068 88.011 1.839 † 40 Phyllastrephus icterinus NHMO BI 44618 28/12/2011 Cameroon 12.353 65.919 72.269 84.622 1.762 41 † Phyllastrephus icterinus NHMO BI 83541 01/11/2013 Cameroon 11.804 62.257 72.588 84.393 2.089 42 43 44 45 46 Biological Journal of the Linnean Society 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Biological Journal of the Linnean Society Page 36 of 40
1 2 3 4 5 Table S2. Continued 6 Sperm head Sperm midpiece Sperm flagellum Sperm total 7 Species Accession no Collection date Country length length length length CV wm 8 Phyllastrephus xavieri NHMO BI 83548 01/11/2013 Cameroon 11.741 66.835 76.598 88.339 1.385 9 01/11/2013 10 Phyllastrephus xavieri NHMO BI 83549 Cameroon 12.252 66.274 76.404 88.656 1.756 11 Phyllastrephus xavieri NHMO BI 83550 01/11/2013 Cameroon 12.104 67.855 79.838 91.942 2.346 12 Phyllastrephus xavieri NHMO BI 83551 01/11/2013 Cameroon 11.201 65.552 72.219 83.420 1.528 13 Pycnonotus barbatus NHMO BI 25691For 30/04/2010 Peer Nigeria Review 14.09 62.323 68.97 83.06 1.722 14 Pycnonotus barbatus NHMO BI 25799 13/06/2010 Nigeria 12.662 58.593 67.964 80.626 2.628 15 16 Pycnonotus barbatus NHMO BI 43644 17/03/2011 Nigeria 13.951 57.907 73.031 86.983 2.943 17 Pycnonotus barbatus NHMO BI 43645 07/06/2011 Nigeria 13.864 59.426 71.78 85.644 1.811 18 Pycnonotus barbatus NHMO BI 43676 14/07/2011 Nigeria 13.308 63.915 73.735 87.043 2.897 19 Pycnonotus barbatus NHMO BI 32585 25/06/2012 Nigeria 13.85 61.281 71.988 85.838 1.696 20 Pycnonotus barbatus NHMO BI 32908 03/07/2012 Nigeria 13.159 61.378 72.371 85.531 1.674 21 Pycnonotus barbatus NHMO BI 32943 04/07/2012 Nigeria 14.45 60.099 74.382 88.833 2.408 22 23 Pycnonotus barbatus NHMO BI 33248 18/07/2012 Nigeria 11.968 62.431 72.595 84.563 1.479 24 Pycnonotus barbatus NHMO BI 33250 18/07/2012 Nigeria 12.972 59.476 71.089 84.062 3.092 25 Pycnonotus barbatus NHMO BI 33306 19/07/2012 Nigeria 13.128 57.697 69.212 82.341 2.640 26 Pycnonotus barbatus NHMO BI 33347 20/07/2012 Nigeria 12.55 55.861 63.081 75.63 2.592 27 19/04/2013 28 Pycnonotus barbatus NHMO BI 35649 Nigeria 13.616 57.538 68.319 81.936 2.064 29 Pycnonotus barbatus NHMO BI 35680 22/04/2013 Nigeria 13.876 60.437 67.091 80.967 1.858 30 Pycnonotus barbatus NHMO BI 46229 26/04/2013 Nigeria 14.357 60.361 70.301 84.658 2.433 31 Pycnonotus barbatus NHMO BI 46230 29/04/2013 Nigeria 13.033 58.741 68.118 81.15 3.762 32 Pycnonotus barbatus NHMO BI 35772 29/04/2013 Nigeria 13.91 61.099 72.67 86.58 1.588 33 29/04/2013 34 Pycnonotus barbatus NHMO BI 35780 Nigeria 13.952 61.461 70.359 84.311 2.06 0 35 Pycnonotus barbatus NHMO BI 35818 01/05/2013 Nigeria 13.474 58.262 65.911 79.385 2.221 36 Pycnonotus barbatus NHMO BI 46240 03/07/2013 Nigeria 13.238 58.202 71.281 84.519 3.787 37 Arizelocichla montana NHMO BI 46303 03/12/2008 Cameroon 13.325 59.888 71.592 84.916 1.547 38 Arizelocichla montana NHMO BI 26187 18/12/2010 Cameroon 13.96 60.381 71.893 85.853 2.346 39 40 Arizelocichla montana NHMO BI 26220 18/12/2010 Cameroon 13.25 62.609 71.759 85.009 2.443 41 Arizelocichla montana NHMO BI 46305 30/12/2008 Cameroon 13.462 59.866 72.645 86.107 2.576 42 43 44 45 46 Biological Journal of the Linnean Society 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 37 of 40 Biological Journal of the Linnean Society
1 2 3 4 5 6 Table. S2 Continued 7 Sperm head Sperm midpiece Sperm flagellum Sperm total 8 Species Accession no Collection date Country length length length length CV wm 9 Arizelocichla montana NHMO BI 26273 18/12/2010 Cameroon 13.822 66.987 73.957 87.78 1.115 10 11 Arizelocichla montana NHMO BI 26317 18/12/2010 Cameroon 14.555 61.248 75.096 89.651 1.348 12 Arizelocichla montana NHMO BI 26364 18/12/2010 Cameroon 13.742 61.58 72.763 86.505 2.765 13 Arizelocichla tephrolaema NHMO BI 26112For Peer18/12/2010 Cameroon Review 13.589 44.553 54.177 67.765 2.165 14 Arizelocichla tephrolaema NHMO BI 26235 20/12/2010 Cameroon 14.192 46.94 59.731 73.924 4.644 15 Arizelocichla tephrolaema NHMO BI 26308 18/12/2010 Cameroon 13.387 43.574 55.702 69.089 3.199 16 17 Arizelocichla tephrolaema NHMO BI 26309 18/12/2010 Cameroon 12.618 46.216 57.113 69.732 3.215 18 Arizelocichla tephrolaema NHMO BI 26318 18/12/2010 Cameroon 13.933 46.655 57.799 71.732 3.128 19 Eurillas ansorgei NHMO BI 46023 18/12/2010 Cameroon 13.4 82.521 94.987 108.387 2.700 20 Criniger chloronotus NHMO BI 46109 18/12/2010 Cameroon 15.494 51.815 71.375 86.869 3.236 21 Phyllastrephus poensis NHMO BI 45890 19/12/2010 Cameroon 12.807 55.701 66.485 79.292 2.111 22 23 Phyllastrephus poliocephalus NHMO BI 45901 19/12/2010 Cameroon 12.408 60.643 68.62 81.028 1.772 24 Phyllastrephus poensis NHMO BI 45907 19/12/2010 Cameroon 12.36 59.139 68.561 80.921 2.125 25 Phyllastrephus poensis NHM O BI 45949 19/12/2010 Cameroon 12.344 59.446 66.825 79.169 1.939 26 27 † Individual not included in ANOVA and PGLS analysis 28 NHMO BI = Natural History Museum Oslo – Bird Collection 29 30 31 32 33 34 35
36
37
38
39
40 41 42 43 44 45 46 Biological Journal of the Linnean Society 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Biological Journal of the Linnean Society Page 38 of 40
1 2 Table S3. Regression analysis controlling for phylogeny (PGLS) among sperm traits in greenbuls (N = 21 3 4 species) and one bulbul. The model including the maximum likelihood of lambda ( λ) value was compared 5 6 against the models including λ = 1 and λ = 0, and superscripts following the λ values indicate probability 7 8 (P) of likelihood ratio of sperm trait (first position: against λ = 0; second position: against λ = 1) 9 10 11 Sperm traits β ± SE t P λ 12 13 Head and midpiece 0.03 ± 0.01 5.24 <0.0001 10.005; 1.00 14 15 Head and flagellum 0.02 ± 0.02 1.22 0.24 0.800.002; 0.006 16 17 Head and total length 0. 03 ± 0. 02 1. 62 0.1 2 0. 780.003; 0.001 18 For Peer Review 19 Midpiece and flagellum 1.07 ± 0. 06 17.66 <0.000 1 0.37 0.48; <0.001 20
21 0.31; <0.001 22 Midpiece and total length 1.02 ± 0.06 15.91 <0.0001 0.39 23 0.01; <0.01 24 Flagellum and total length 0.95 ± 0.02 53.94 <0.0001 0.71 25 1.00; <0.001 26 Head and midpiece:flagellum 0. 03 ± 0. 07 0.40 0. 69 0 27 28 Midpiece:flagellum and total length 0.02 ± 0. 01 3.38 <0.00 3 0.37 0.49; <0.001 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Biological Journal of the Linnean Society Page 39 of 40 Biological Journal of the Linnean Society
1 2 Table S4. Regression analysis controlling for phylogeny (PGLS) between sperm traits and sperm 3 4 competition (as CV bm ) among 10 greenbul and one bulbul species. The model including the maximum 5 6 likelihood of lambda ( λ) value was compared against the models including λ = 1 and λ = 0, and superscripts 7 8 following the λ values indicate probability ( P) of likelihood ratio of sperm trait (first position: against λ 9 10 = 0; second position: against λ = 1) 11 12 13 14 Sperm traits β ± SE t P λ 15 16 Head 0.06 ± 0.22 0.25 0.80 10.002; 1.00
17 18 Midpiece For Peer 0.95 ± 2.68 Review 0.35 0.73 10.02; 1.00 19 20 Flagellum 0.91 ± 2.66 0.34 0.74 10.02; 1.00 21 22 Total length 0.96 ± 2.79 0.34 0.74 10.02; 1.00 23
24 Flagellum:head 0.05 ± 0.16 0.33 0.75 10.01; 1.00 25 26 Midpiece:total length 0.01 ± 0.04 0.24 0.81 10.0 2; 1.00 27 28 Midpiece:flagellum 0.01 ± 0.05 0.26 0.80 10.06; 1.00 29 30
31 32 33 34 35 36 37
38 39 40 41 42 43 44
45 46 47 48 49 50 51
52 53 54 55 56 57 58 59 60 Biological Journal of the Linnean Society Biological Journal of the Linnean Society Page 40 of 40
1 2 REFERENCES 3 4 BirdLife International. 2012. The BirdLife checklist of the birds of the world, with conservation status 5 6 and taxonomic sources. Version 5.1 . 7 Gill F, Donsker D. 2015. IOC World Bird List version 4.2. Available at: www.worldbirdnames.org. 8 9 Hebert PDN, Stoeckle M Y, Zemlak T S, Francis CM. 2004. Identification of birds through DNA 10 barcodes. Plos Biology 2: 1657–1663. 11 12 Jetz W, Thomas GH, Joy JB, Hartmann K, Mooers AO. 2012. The global diversity of birds in space 13 and time. Nature 491: 444–448. 14 15 John H, Boyd III. 2013. TiF checklist, Version 2.87 . 16 Johnsen A, Rindal E, Ericson P, Zuccon D, Kerr K, Stoeckle M, Lifjeld J. 2010. DNA barcoding of 17 18 ScandinavianFor birds reveals Peer divergent lineages Review in trans Atlantic species. Journal of Ornithology 19 20 151: 565–578. 21 Lohman DJ, Prawiradilaga DM, Meier R. 2008. Improved COI barcoding primers for Southeast Asian 22 23 perching birds (Aves: Passeriformes). Molecular Ecology Resources 9999. 24 Ratnasingham S, Hebert PDN. 2007. BOLD: The Barcode of Life Data System. Availaible at: 25 26 www.barcodinglife.org. Molecular Ecology Notes 7: 355–364. 27 Tamura K, Stecher G, Peterson D, Filipski A, Kumar S. 2013. MEGA6: Molecular Evolutionary 28 29 Genetics Analysis version 6.0. Molecular Biology and Evolution 30: 2725–2729. 30 The Internet Bird Collection. 2012. Ibc.lynxeds.com. 8.http://ibc.lynxeds.com/. 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Biological Journal of the Linnean Society