Genetic and Morphological Assessments of Hybridization Between Two Non-Native Geoemydid Turtles, Mauremys reevesii and Mauremys mutica, in Northcentral Japan

• Published in: Chelonian conservation and biology Vol. 13; no. 2; pp. 191 - 201
• Main Authors: Fujii, Ryo, Ota, Hidetoshi, Toda, Mamoru
• Format: Journal Article
• Language: English
• Published: Department of Psychology, Indiana University, Bloomington, Indiana 47405 The Society for the Experimental Analysis of Behavior, Inc 01.12.2014; Allen Press Inc
• Subjects: alleles; allozyme; allozymes; backcross; backcrossing; Chinemys reevesii; Evolution; genotype; Genotypes; Hybridization; Hybrids; interspecific hybridization; introgression; islands; Japan; loci; Mauremys; Mauremys mutica; Morphology; nucleotide sequences; Ponds; Population; random mating; Reptiles & amphibians; s; SINE; turtle; Turtles; wildlife management; Japan; Japan, Honshu; INW, Japan, Honshu
• ISSN: 1071-8443; 1943-3956
• DOI: 10.2744/CCB-1067.1

Interspecific hybridization has been occasionally reported for various combinations of geoemydid turtles, but genetic consequences of such hybridization have rarely been investigated. We surveyed the hybrid status of 40 individuals from a seemingly hybridizing turtle assemblage in a natural pond in northcentral Honshu, Japan. Analysis of mtDNA sequences confirmed that the parental species of this assemblage were Mauremys reevesii (Chinese three-keeled pond turtle) and Mauremys mutica (Asian yellow pond turtle), neither of which is native to the main islands of Japan, although M. reevesii was naturalized in the area before the 19th century. Extensive examination of pure strain samples of the 2 species yielded 10 morphological and 10 genetic diagnostic characters (allozyme and short interspersed repetitive elements) that were used to determine the hybrid status of the turtles in the assemblage. Morphological examination showed that 19 out of 40 individuals were not different from pure M. reevesii individuals, but the remaining 21 individuals exhibited M. mutica states at 1 to 6 characters, indicating interspecific crosses at a moderate to large scale. In the genetic characters, however, only 2 of the 21 individuals possessed marker alleles of M. mutica at a few loci and all others possessed alleles of M. reevesii only; thus neither pure M. mutica nor F1 hybrids were included in the assemblage. These results suggest that iterative backcrosses with pure M. reevesii have occurred. In the analysis using the genetic hybrid index (GHI), which was defined as the number of M. mutica type alleles over the 10 examined loci, the 2 individuals scored 5 and 2, respectively, and all others scored 0. This discrete distribution of GHI scores negates an assumption of a random mating hybrid swarm but suggests that these two individuals emerged by a different hybridization event, independent from that leading to the many other putative hybrids detected by morphological characters only. Our results suggest that even though the hybrid genotypes rarely prevailed in the population, once introgressed genes were diluted by successive backcrosses, they could have been retained over many generations.