Background: Noninvasive and nondestructive DNA sampling techniques are becoming more important in genetic studies because they can provide genetic material from wild animals with less or even without disturbance,which is particularly useful for the study of endangered species,i.e.,birds.However,nondestructively and noninvasively sampled DNA may,in some cases,be inadequate in the amount and quality of the material collected,which can lead to low amplification success rates and high genotyping errors.Methods: In this study,noninvasive(eggshell swab,shed feather and feces),nondestructive(plucked feather and buccal swab) and invasive(blood) DNA samples were collected from the vulnerable Chinese Egret(Egretta eulophotes).DNA concentrations,PCR amplification success and microsatellite genotyping errors of different sample types were evaluated and compared to determine whether noninvasive and nondestructive samples performed as well as invasive samples in our experimental procedures.Results: A total of 159 samples were collected in the field.Among the different sample types,the highest DNA concentrations(154.0–385.5 ng/μL) were obtained from blood.Those extracted from fecal samples were the lowest,ranging from 1.25 to 27.5 ng/μL.Almost all of the DNA samples,i.e.,95.59 %,were successfully amplified for mt DNA(n = 152) and 92.76 % of mt DNA samples were successfully genotyped for at least five of the nine microsatellite loci tested(n = 141).Blood samples and buccal swabs produced reliable genotypes with no genotyping errors,but in feces,allelic dropouts and false alleles occurred in all nine loci,with error rates ranging from 6.67 to 38.10 % for the dropouts and from 6.06 to 15.15 % for the false alleles.Conclusions: These results indicate that both nondestructive and noninvasive samplings are suitable for avian microsatellite genotyping,save for fecal DNA.However,we should remain cautious of the appearance of genotyping errors,especially when using noninvasive material.
