Study animal
B. sitibundus is widely distributed across a broad range extending from Greece eastwards through Turkey, Syria, Jordan, and Lebanon, and further into Iraq and Iran. This species spans the Caucasus and Russia, reaching Kazakhstan, with isolated populations reported in Denmark, northern Germany, and southern Sweden (Fakharzadeh & Hosseinzadeh 2021). Historically classified as Bufotes viridis with three subspecies, recent molecular studies have redefined the taxon, recognizing B. sitibundus as a distinct species. In Iran, B. sitibundus is primarily found in the western and central regions, previously identified under B. viridis with subspecies distributed in different provinces. This species is a member of the B. viridis complex, which is notable as the only known anuran group comprising diploid, tetraploid, and triploid taxa that reproduce bisexually and are widespread throughout the Palearctic. In Iran, all three ploidy levels (2n, 3n, 4n) have been documented, although cytogenetic evidence from several populations indicates that B. sitibundus in this region is predominantly diploid (Fakharzadeh et al. 2015).
Sampling
Kazzaz (49◦26'15''E and 34◦00'10''N) is a village in Pol-e Doab Rural District, Zalian District, Shazand County, Markazi Province, Iran. At the 2006 census, its population was 2,148, in 536 families. In April 2022, we collected B. sitibundus egg clutches (3500 eggs from the same clutches) from one of the B. sitibundus breeding habitats (under the permission of the Ethics Committee of Arak University (IR.ARAKU.REC.1401.080) that were free of any butachlor herbicide contamination. To avoid any influence from pesticides, the eggs were obtained from a spring with clean water in an area where there are no agricultural lands or evidence for herbicides in this area. We collected the eggs right after the deposition and immediately transferred them to the lab.
Herbicide concentrations
We based the butachlor concentrations on the environmentally relevant concentrations previously reported in the environment. Kaur et al. (2017) reported the range of butachlor concentrations from 0.121 ± 0.051 µg/L to 0.357 ± 0.338 µg/L in the water of rice farms after herbicide spraying. Concentrations of butachlor reported to be lethal to 50% of amphibian embryos and larvae (LC50s) range from 0.53 mg/L (96 h, Microhyla ornata) to 2.67 mg/L (24 h, Polypedates megacephalus) (Abigail et al. 2014; Li et al. 2016). We therefore used a series of butachlor concentrations of 1.5, 1, 0.8, 0.6, 0.4, and 0.1 mg/L, plus a control group with no added herbicide.
To prepare the desired concentrations of butachlor for the experiment, we first obtained emulsion butachlor with a purity of 65% (CAS No. 23184-66-9) from (Behsam Alborz). A stock solution of butachlor was prepared by dissolving the required amount of the herbicide in deionized water to achieve a stock concentration. The stock solution was then serially diluted using deionized water to obtain the working concentrations needed for the experiment. We stored the stock solution in a dark, airtight glass container at 4°C throughout the incubation period to prevent degradation and maintain stability. Fresh working solutions were prepared weekly by diluting the stock solution immediately before use to ensure the accuracy and consistency of the herbicide concentrations in the experimental setups. All glassware and containers used in the preparation and storage of butachlor solutions were thoroughly cleaned and rinsed with deionized water to avoid contamination.
Experimental design
The green toad’s eggs were housed in seven polyethylene tubes, each containing 500 eggs and approximately 5L of water. Six of the tubes contained different concentrations of butachlor: 1.5 mg/L, 1 mg/L, 0.8 mg/L, 0.6 mg/L, 0.4 mg/L, and 0.1 mg/L, while the seventh tube, serving as the control group, contained clean habitat water. We started the incubation period by adding the prepared concentrations to each tube. The water in each tube was renewed every seven days during the experiment, and a new stock of herbicide concentrations was added to each treatment. To ensure consistent oxygen levels across all treatments during the incubation period, oxygen was continuously supplied to the rearing tubes using an air pump. Since the rearing area was in an open space with natural light and subject to the ambient light and dark cycles, we did not interfere with these conditions. Additionally, the temperature and humidity of the environment, as well as the temperature of the rearing containers, were recorded throughout the entire incubation period (see Fig. S1-S3). To maximize the similarity of the study environment to the natural breeding habitat, the study was conducted in an open environment influenced by natural temperature and light-dark cycles. Temperature and humidity fluctuations during the day and night were recorded throughout the entire incubation period.
The incubation period continued until the completion of the metamorphosis of all individuals. Given that the development of the oral structure of larvae is completed at Gosner stage 25 (Gosner 1960), cooked lettuce leaves, boiled potatoes, and cooked spinach were used to feed the larvae at this stage (two times/day).
Developmental stage parameters
We measured the growth, death rates, length at the beginning of metamorphosis (LBM (mm)), length at the formation of forelimb bud (LFF (mm)), length at the end of metamorphosis (LEM (mm)), weight at the beginning of metamorphosis (WBM (gr)), weight at the end of metamorphosis (WEM (gr)) for a sample size of three in each phase (Gosner, 1960).
Parameters such as the length of the incubation period, the number of metamorphoses, the number of eggs that perished before entering the embryonic stage, the number of embryos that perished before entering the larval stage, and the number of larvae that did not complete metamorphosis were compared in different groups during the incubation period.
Data analyses
We compared the length of the incubation period, the number of metamorphoses, the number of dead eggs before entering the embryonic stage, the number of dead embryos before entering the larval stage, and the number of larvae that did not complete metamorphosis across different treatments and the control group. Survival analyses were conducted using the Kaplan-Meier (1958) method, a non-parametric statistic used to estimate the survival function from lifetime data. This method is particularly useful for handling censored data, such as when a subject leaves the study before an event occurs. The Kaplan-Meier survival curves, which display survival probabilities over time as a series of declining horizontal steps, were compared across different concentrations using the Mantel-Haenszel log-rank test (Harrington & Fleming 1982). This test compares the survival distributions of two or more groups by testing the null hypothesis that there is no difference in survival experiences across the groups.
We also used Multivariate Analysis of Variance (MANOVA) and Univariate Analysis of Variance (ANOVA) to compare the length and weight of metamorphosed larvae (LBM, LFF, LEM, WBM, WEM) across different treatment groups. MANOVA was employed to assess multiple dependent variables simultaneously, particularly useful when these variables are correlated. Eta squared was calculated as an effect size measure to quantify the proportion of variance in the dependent variables attributable to the independent variable(s), with values around 0.01 and 0.06 indicating small to medium effects, and values greater than 0.14 indicating large effects. Post-hoc comparisons were made using the Tukey HSD (Honestly Significant Difference) test to determine which specific group means differed significantly from each other in the various butachlor treatments and the control.