Deborah Duffield

Date of Award

Fall 12-10-2014

Document Type


Degree Name

Doctor of Philosophy (Ph.D.) in Biology



Physical Description

1 online resource (xii, 173 pages)


Giant panda -- Breeding -- Conservation -- China, Giant panda -- China -- Reproduction, Captive mammals -- Breeding, Sexual selection in animals, Courtship in animals




Successful captive-breeding and re-introduction programs must have the ability to breed a surplus of genetically suitable animals for release into the wild. Unfortunately, many individuals in captive breeding programs often do not reproduce even when they are apparently healthy and presented with genetically appropriate mates. Mate choice can affect multiple parameters of reproductive sperformance, including mating success, offspring production, survival, and fecundity. We investigated the role of mate preference and personality on the reproductive performance of male and female giant pandas (Ailuropoda melanoleuca) as measured by intromission success and litter production. We conducted these studies on giant pandas at the China Conservation and Research Center for the Giant Panda (Bifengxia, Sichuan, China.) from February-May 2011, 2012, and 2013.

In Chapter 2, we compared reproductive performance between individuals mated with preferred and nonpreferred males of the focal female (female preference), the focal male (male preference), and mutual mate choice perspective. There were four types of reproductive pairings in our mutual mate choice analyses: females and males that preferred each other (P-P), females that preferred the male but the male did not prefer the female (P-NP), females that did not prefer the male but the male preferred the female (NP-P), and females and males that did not prefer each other (NP-NP). Pairing giant pandas with a preferred partner increased the probability of intromission success and producing a cub. Of the 25 females mated to a preferred partner, 72% had successful intromission, 64% produced cubs, and 52% reared their own cubs versus 31%, 12%, 12% for females mated to their nonpreferred partner (N = 16). Male giant panda mate preference showed similar results. Of the 24 males that were mated with their preferred females, 75% had successful intromission, 77.8% produced cubs, and 92.9% produced cubs that were maternally reared versus 31%, 60%, 66.7% for males mated with their nonpreferred females (N = 16). There was a statistically significant difference in intromission success and cub production for P-P pairings versus NP-NP pairings. Of the twelve P-P mate dyads 83% had successful intromission, 90% produced cubs, and 88.9% had mothers rear their cubs. Of the four P-NP mate dyads 50% had successful intromission, 50% produced cubs, and 100% had the mother rear their cub. Of the seven NP-P mate dyads 57% had successful intromission, 50% produced cubs, and 100% had the mothers rear their cubs. Of the three NP-NP mate dyads 0% had successful intromission resulting in 0% producing cubs and having mother-rear cubs. In addition, we took into account various life and breeding history factors that may have affected reproductive performance such as birth location, rearing, previous sexual history with the opposite sex and familiarity status directly prior to breeding. Mate pairings that were familiar with each other as measured by successfully producing a cub in the past, had increased intromission success. Females mated to heavier males had more intromissions and produced more cubs than females mated to smaller males.

In Chapter 3 and 4, we investigated the effect of personality on reproductive performance measurements. Recently, personality of individual animals has been implicated in the failure or success of captive management and breeding of threatened and endangered species. Measurements of personality included novel object tests (where behaviors of individual giant pandas were scored toward a novel object for one hour) and keeper surveys (where the primary keeper rated giant pandas on 23 adjectives describing personality); both measures significantly correlated with variation in long-term and specific mating dyad reproductive performance.

In Chapter 3 we evaluated the effect of personality on long-term reproductive performance. Our results show that for both male and female giant pandas, `fearful' individuals had lower natural mating success and cub production. In addition, `aggressive' female giant pandas had increased natural mating success but decreased cub production. Males did not show these relationships between `aggressive' personality types and reproductive performance. Female giant pandas that were labeled as `aggressive' had higher frequencies of positive mating behaviors but `fearful of conspecifics' females had lower frequencies. Males labeled `fearful of conspecifics' and with high `playful-clever' component scores had a lower frequency of positive mating behaviors.

In Chapter 4 we further investigated the effects of personality within a specific mating dyad. Recent studies suggest that mate compatibility within personality types can improve reproductive performance measurements such as mating success, offspring survivorship, and offspring quality. We investigated the effect of personality compatibility on reproductive performance through analyzing mate dyad similarity in personality (similar versus dissimilar), specific personality combinations, and relative personality scores (male scores higher or lower than female scores). The specific personality combinations were: females and males that were both low on the personality component (L_L), females were low but males were high (L_H), females were high but males were low (H_L), and females and males were both high (H_H). Our results suggest that giant pandas appear to be assortatively mating based on aggressive, fearful, and stereotypical-excitable personality traits. Mate dyads that were dissimilar on the stereotypical-excitable component score were more likely to have successful intromission than pairs that were similar. L_H mate dyads on the aggressive and stereotypical-excitable personality traits had increased intromission success and cub production than all other possible personality combinations. L_L on fearfulness had higher intromission success but not higher cub production than all other possible personality combinations. Mate dyads composed of males ranked high on the aggressive component relative to the female resulted in higher reproductive performance and cub production.

In captive breeding programs, mates are traditionally selected solely on the basis of genetic parameters to minimize loss of genetic diversity and inbreeding coefficients. Our results suggest that integrating genetic with behavioral measures of preference, breeding history factors, and personality would increase the reproductive output of the giant panda captive-breeding program. We stress the importance of considering husbandry practices to accommodate personality traits and ultimately increase reproductive performance in the captive population while simultaneously maintaining both genetic and behavioral diversity.

Persistent Identifier