How to do time comparisons on repeated measures in R? I repeatedly measured one variable on three replicates of one subject at successive time points. The data is as follow:
df <- data.frame(time = c(10,20,30,50,100,200,500,1000),
                    A = c(1.5,2.5,2.8,4.5,5.4,6.2,7.4,8.1),
                    B = c(1.6,2.6,2.7,4.4,4.9,6.1,7.5,8),
                    C = c(1.4,2.4,2.6,4.2,4.9,5.9,7.8,7.9))

Now I would like to know if the variable values are significantly different among these times? Because these measurements are obtained for the same units, they are auto-correlation along the time; the ordinary "Least signiﬁcant difference" in R package "agricolae" may be not suitable for this task.  Could anybody help me to solve this problem?
A further question, if I have multiple subjects and three replicates for each subject, then how can I compare their means along the time? Thanks in advance! 
 A: What I think you want to do is just a regular ANOVA.
First I melted your data to use as input for the ANOVA
melted_df <- data.table::melt(df, id.vars = c("time"), variable.name = c("group"))

Next I want to perform an ANOVA which takes into account both time and the sample group (either A, B or C) and also the interaction between them: group:time. 
 av <- aov(value~group+time+group:time, data = melted_df)
 av

        Df Sum Sq Mean Sq F value   Pr(>F)    
 group        2   0.11    0.05   0.025    0.975    
 time         1  83.26   83.26  39.506 6.32e-06 ***
 group:time   2   0.02    0.01   0.005    0.995    
 Residuals   18  37.93    2.11                     
 ---
 Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

This shows that only "Time" has a significant effect on the results.
HOWEVER: I think you should not use repeated measures in the first place, depending on what kind of data you have, you can either make independent measurements or try to combine the measurements per sample into 1 value. 
