While statistical significance calculus is not my expertise, my laboratory neuroscience studies speak to the importance in writing up experimental designs that best determine the statistical path you take and must be hypothesis-based. That said, I am unclear re what is your statement of hypothesis.
After a fast read of your data, the first question to be asked and answered is what is the power calculation of a normal distribution for a specific n number of groups of equal variances with a mean population for example of 1=100, or 2=130, sigma=10, p value level = 0.05. power=0.8, n number of sides=2. The results of this power calculation will inform you of the n number of data points/sets required in each grouping of an ANOVA. Of note, changing the sigma is positively correlated to the data points/sets required in each group. Posthoc tests certainly should be done on all types of ANOVAs, designating specific contrasts vs all possible differences. I hope this is helpful….
