名古屋地方における蚊族個体群の発生動態 : IV. 野外地域で捕集したアカイエカ群成虫の判別関数法による形態学的検討

Trial application of discriminant function was proposed for the differentiation of three types of Culex pipiens complex, C.p. fatigans Wiedemann, 1828 (Cpf), C.p. pallens Coquillett, 1898 (Cpp) and C.p. molestus Forskal, 1778 (Cpm) in Japan. Using this method, a new approach was made to the problem whether Cpm or Cpf adults were mixed or not in "Cpp imaginal populations" captured in the open field of Nagoya City, central Japan. 1) The mean values of diagnostic characters in the specimens of Cpf (Naha strain). Cpp (Nagoya strain) and Cpm (Nagoya strain) used as a temporary standard for comparison were as follows respectively : D/V ratio of male genitalia (Fig. 2)=0.38,0.79 and 1.08; SB value (Fig. 2)=17.1,11.1 and 5.8 unit (1 unit=2.5μ); ommatidial number of female (Fig. 3)=80.0 (8.9/row), 78.6 (8.9/row) and 60.7 (7.8/row); HS value of female genitalia (Fig. 3)=22.5,21.3 and 25.6 unit (Table 1 and Figs. 2,3). 2) When x_1 denotes the D/V ratio, x_2 the SB value of male, and x_1' denotes the ommatidial number (square value of the average number per row), x_2' the HS value of female, the discriminant functions are determined as general equations, Z_m=λ_1x_1+λ_2x_2 and Z_f=λ_1x_1'+λ_2x_2' (Table 2). The discriminant values in all of the specimens were calculated by substituting corresponding values of the above characters for x_1 or x_1' and x_2 or x_2' in the respective equation. As a result, frequency distributions of these discriminant values were clearly separated from each other with only a few exceptions (Figs. 4,5). 3) The critical point of the discriminant values was fixed so that α might be equal to β for male mosquitoes, where α signified the probability of committing type I error in which Cpp might be mistaken for Cpm (or Cpf) and β signified that of type II error in which Cpm (or Cpf) might be mistaken for Cpp (Fig. 4). The rejection limit of the discriminant values of Cpp (one sided probability of type I error, α=0.01) was used as a critical point in differentiating Cpm from Cpp in the case of famale (Fig. 5). 4) Judging by the above critical discriminant values, the number of male mosquitoes morphologically identified as Cpm and Cpf was 0 and 2 (ca. 0.5%) respectively, and the number of female identified as Cpm was 2 (ca. 2%) in the Cpp imaginal populations captured in the open field (Figs. 6,7,8 and Table 3). 5) An advantage was discussed that an integrated and objective discrimination could be expected when the discriminant function derived from the quantitative diagnostic characters is properly applied to the identification of allied species. A discussion was given of supplementary use of the function as an aid to differentiate only indistinguishable individuals by one character.