I examined fission-fusion sociality in dusky dolphins (Lagenorhynchus obscurus), and investigated aspects of social convergence between dolphins and great apes. I used boat-based group focal follows and photo-identification to collect data in Admiralty Bay, New Zealand during 2005-2006. I used generalized estimating equations to examine relationships between party (group) size, rate of party fission-fusion, activity, and location; and relationships between leaping frequency and behavior. Using photo-identification images from 2001-2006, I analyzed the strength and temporal patterning of associations, short- and long-term association patterns, preferred/avoided associations, and behaviorally-specific preferred associations. To analyze social convergence between dolphins and great apes, I compared female bottlenose dolphin (Tursiops spp.) and chimpanzee (Pan troglodytes) social strategies through literature review. I conducted 171 group focal follows, totaling 157 observation hours. Mean party size was 7.0±6.0 individuals. Party size changed every 5±.47.6 min on average. The most frequent activity was resting (37%), followed by traveling (29%), foraging (18%), and socializing (15%). Foraging was positively related to party size and rate of fission-fusion. Near mussel farms, foraging increased, traveling decreased, and rate of party fusion increased. "Clean" leaps were the most frequent leap type (84%) and were positively related to party size and foraging. Noisy and coordinated leaps were positively related to party size; noisy leaps were negatively related to foraging. Associations during 2001-2006 (N = 228 individuals) were nonrandom for 125 days; associations within one field season were nonrandom for 60 days. Individuals formed preferred/avoided associations during most years. The strongest associations occurred during foraging and socializing; the weakest associations occurred during traveling. Individuals formed preferred associations during foraging, resting, and socializing. Review of female bottlenose dolphin and chimpanzee sociality revealed that: 1) females form weaker bonds and are less social than males, 2) females associate mostly with other females, 3) mothers are often alone with their offspring, 4) mothers (vs. non-mothers) and non-cycling (vs. cycling) females associate less with males, and 5) non-cycling (vs. cycling) females occur in smaller parties. Female dolphins may be more social than female chimpanzees due to decreased scramble competition, increased predation risk, and decreased cost of transport for dolphins vs. chimpanzees.

Great ape gesture is an elaborate, flexible system of intentional communication. It has been suggested that human language originated in gesture, thus, the gestural communication of great apes is of great interest for questions on the origin of language. To date, systematic studies of great ape gesture have been limited to restricted captive settings, supplemented by the study of a few specific gestures in wild populations. To address questions about gestural communication from an evolutionary perspective it is necessary to extend the systematic study of gesture into a wild ape population. I therefore undertook a 22-month study of gesture in the wild Sonso chimpanzee community in Budongo, Uganda. Sonso chimpanzees employ a large repertoire of species-typical gestures in intentional communication; a proportion of this repertoire appears to be ape-typical, as would be expected with a biologically given trait. Chimpanzees can acquire new behavioural patterns through imitation; however, this apparently does not represent a significant means of acquiring gestures. Gesturing was employed regularly in an intentional manner from the end of the first year, and was used by chimpanzees of all ages to communicate across a range of contexts, including the evolutionarily urgent context of consortship. Immature chimpanzees used a wide range of gestures, which they combined into rapid sequences. With maturity, use of the repertoire was ‘tuned’ to focus on the most effective gestures, which were then used individually. Despite the evidence for referential pointing in captive chimpanzees, there was little evidence for the regular use of it in wild chimpanzees. Gestures were used to communicate a range of imperative requests that regulated social behaviour. Chimpanzee gestures vary from the ambiguous to the highly specific in meaning; and, while gestures were used flexibly, they tended to be associated with a single dominant meaning.

The mitochondrial control region (MCR) has played an important role as a population genetic marker in many taxa but sequencing of complete eukaryotic genomes has revealed that nuclear integrations of mitochondrial DNA (numts) are abundant and widespread across many taxa. If left undetected, numts can inflate mitochondrial diversity and mislead interpretation of phylogenetic relationships. Comparative analyses of complete genomes in humans, orangutans and chimpanzees, and preliminary studies in gorillas have revealed high numt prevalence in great apes, but rigorous comparative analyses across taxa have been lacking. The present study aimed to systematically compare the evolutionary dynamics of MCR numts in great apes. Firstly, an inventory numts derived from the region containing the MCR subdomains was carried out by genomic BLAST searches. Secondly, presence/absence of each candidate numt was determined in great ape taxa to estimate numt insertion rate. Thirdly, alternative mechanisms of numt insertion, either through direct mitochondrial integration or post-insertional duplications, were also assessed. Fourthly, the effect of nuclear and mitochondrial environment on patterns of nucleotide composition and substitution was assessed through sequence comparisons of nuclear and mitochondrial paralogous sequences. Finally, numts in the gorilla genome were identified through two experimental methods and their use as polymorphic genetic markers was then evaluated in a sample of captive gorillas from U.S. zoos. A deficit of MCR numts covering two particular mitochondrial subdomains was detected in all three apes examined, and is largely attributed to rapid loss of mitochondrial and nuclear sequence identity in the mitochondrial genome. Insertion rates have varied during the great ape evolution and exhibit substantial differences even between related taxa. The most likely mechanism of numt insertion is direct mitochondrial integration through Non-Homologous-End-Joining Repair. Transition/transversion ratios differed significantly between both mitochondrial and nuclear sequences and between numts from coding and non-coding mitochondrial regions. A previously documented upward bias in the GC content of the primate mitochondrial genome was confirmed and the extent of this bias relative to the corresponding numt sequences increased with numt age. Five gorilla-specific numts were isolated, including three exhibiting insertional polymorphisms that will be used in future population genetic studies in free-range gorilla.

付記する学位プログラム名: 霊長類学・ワイルドライフサイエンス・リーディング大学院 / 京都大学 / 新制・課程博士 / 博士(理学) / 甲第23054号 / 理博第4731号 / 新制||理||1678(附属図書館) / 京都大学大学院理学研究科生物科学専攻 / (主査)准教授 足立 幾磨, 准教授 後藤 幸織, 教授 高田 昌彦 / 学位規則第4条第1項該当 / Doctor of Science / Kyoto University / DFAM

While most human language is expressed verbally, the gestures produced concurrent to speech provide additional information, help listeners interpret meaning, and provide insight into the cognitive processes of the speaker. Several theories have suggested that gesture played an important, possibly central, role in the evolution of language. Great apes have been shown to use gestures flexibly in different situations and to modify their gestures in response to changing contexts. However, it has not previously been determined whether ape gestures are defined by structural variables, carry meaning, are used to intentionally communicate specific information to others, or can be used strategically to overcome miscommunication. To investigate these questions, I studied three captive populations of orangutans (Pongo pygmaeus and P. abelii) in European zoos for 10 months. Sixty-four different gestures, defined through similarities in structure and use, were included in the study after meeting strict criteria for intentional usage. More than half of the gesture types were found to coincide frequently with specific goals of signallers, and were accordingly identified as having meanings. Both structural and social variables were found to determine gesture meaning. The recipient’s gaze in both the present and the past, and the recipient’s apparent understanding of the signaller’s gestures, affected the strategies orangutans employed in their attempts to communicate when confronted with different types of communicative failure (e.g. not seeing, ignoring, misunderstanding, or rejecting a gesture). Maternal influence affected the object-directed behaviour and gestures of infants, who shared more gestures with their mothers than with other females. These findings demonstrate that gesture can be used as a medium to investigate not only the communication but also the cognition of great apes, and indicate that orangutans are more sensitive to the perceptions and knowledge states of others than previously thought.