抄録
To develop a method for harmlessly identifying individual Japanese charr, Salvelinus leucomaenis, the distribution and shape of parr marks were examined in fluvial specimens. Fish captured in six markandrecapture samples from a tributary of the Tenryu River (35°41'N, 138°07'E, El. 1, 045m) between June 2005 and May 2006 were photographed, and the linear distance and width of each parr mark on the lateral line measured from digital images. Similarities between and within individuals during growth were analyzed using square Euclidean distances. A photomatching test was also conducted.
The number of parr marks and the parr mark ratio increased with body length (BL) up to 50 mm BL. The square Euclidean distance was not correlated with the number of days until recapture, although there was a negative correlation with initial BL (distribution: r=-0.25, P=0.029, width: r=-0.24, P=0.038). The similarity analysis discriminated 74% and 95% of parr mark distribution and width, respectively, in the same individuals. In the photo-matching test, 8 of 10 testers identified 20 individuals from 40 photographs (average accuracy 97%), the exercise requiring about 2 h per tester.
This method, utilizing parr marks, enables the identification of individuals with visible parr marks on the base of tail (BL>32 mm) without harming the fish. This should be useful for identifying fish with parr marks, especially in endangered charr populations.
