Feeding of mussel-associated leeches Hemiclepsis kasmiana on bitterling embryos: Novel interaction between parasites in a shared host

Appendix S1. Environmental conditions of channels

Appendix S2. Statistical analysis

Table S1. Numbers of embryos of bitterling Acheilognathus rhombeus and parasitic leeches Hemiclepsis kasmiana obtained from four mussel species, Nodularia nipponensis, Beringiana fukuharai, Pronodularia cf. japanensis 1, and Obovalis omiensis, placed in a ditch located in the University of Shiga Prefecture. Infestation prevalence (%) denotes the ratio of mussels parasitized by leeches, and infestation intensity denotes the number of leeches per parasitized mussel.

Table S2. Bitterling species compositions in drainage channels I, II, and III where the host choice experiments were conducted for Tanakia limbata, for T. lanceolata, and for Acheilognathus rhombeus, respectively. Surveys were made during the night for intervals including the experiment locations by a hand net (mesh size 3 mm; bl-1L, Mitani-Turigyogu, Saitama, Japan). In channels I, II, and III, the lengths of channel intervals were 90, 100, and 100 m, and the amounts of effort were 40, 80, and 20 min, respectively. Note that the efficiencies of catch per unit effort were different between channels due to the bottom substrates and the presence or absence of obstacles, and among seasons due to precipitation and drainage from paddy fields.

Table S3. Numbers of embryos of bitterling Tanakia limbata and paracitic leeches Hemiclepsis kasmiana obtained from three mussel species, Nodularia nipponensis, Beringiana fukuharai, and Pronodularia cf. japanensis 1, of low-density treatment (without additional infestation of leeches) and high-density treatment (with additional infestation of leeches) in an agricultural drainage channel (Nagahama, Japan). Infestation prevalence (%) denotes the ratio of mussels parasitized by leeches, and infestation intensity denotes the number of leeches per parasitized mussel.

Table S4. Numbers of embryos of bitterling Tanakia lanceolata and parasitic leeches Hemiclepsis kasmiana obtained from three mussel species, Nodularia nipponensis, Beringiana fukuharai, and Pronodularia cf. japanensis 1, of low-density treatment (without additional infestation of leeches) and high-density treatment (with additional infestation of leeches) in an agricultural drainage channel (Nagahama, Japan). Infestation prevalence (%) denotes the ratio of mussels parasitized by leeches, and infestation intensity denotes the number of leeches per parasitized mussel.

Table S5. Numbers of embryos of bitterling Acheilognathus rhombeus and parasitic leeches Hemiclepsis kasmiana obtained from a mussel species Nodularia nipponensis of low-density treatment (without additional infestation of leeches) and high-density treatment (with additional infestation of leeches) in an agricultural drainage channel (Maibara, Japan). Infestation prevalence (%) denotes the ratio of mussels parasitized by leeches, and infestation intensity denotes the number of leeches per parasitized mussel.

Table S6. Results of logistic regression for three bittering species, Tanakia limbata, T. lanceolata, and Acheilognathus rhombeus. Log-odds of the probability that eggs were deposited in a mussel were modeled as a linear model of the mussel shell length and the number of parasitic leeches Hemiclepsis kasmiana inhabited the mussel with intercept $b_0$ and coefficients $b_1$ and $b_2$, respectively. Bayesian inferences were conducted via RStan version 2.21.8 (https://mc-stan.org) on R version 4.2.2 (https://cran.r-project.org) with the settings 4 chains, 1000 iterations, and 500 warmups.

Movie S1. A mussel-associated leech that started feeding on an embryo of bitterling Tanakia limbata that was taken from a dissected host mussel Pronodularia cf. japanensis 1. The leech was still normal brown color.

Movie S2. The same mussel-associated leech that fed on the same embryo as in Supporting Information Movie S1 after 40 min. Color of the leech turned to yellow.