ggplot2 geom_area plots unexpectedly (ends in wrong place)

Question

I'm using ggplot2 to plot the following data using geom_area. The values are designed to be a x axis mirror image. But the end of the plot for dog is not mirrored.

But as you can see from this data frame dog should end on x=100 and y=0 but it seems to end on x=100 and y=100 (see last row)

How can I make dog mirror cat exactly?

       x   y animal
1  100.0 100    cat
2   90.0  89    cat
3   84.0  85    cat
4   55.5  60    cat
5   28.3  37    cat
6   27.0  32    cat
7   18.0  25    cat
8    0.0   0    cat
9    0.0 100    dog
10  10.0  89    dog
11  16.0  85    dog
12  44.5  60    dog
13  71.7  37    dog
14  73.0  32    dog
15  82.0  25    dog
16 100.0   0    dog


dat <- structure(list(x = c(100, 90, 84, 55.5, 28.3, 27, 18, 0, 0, 10, 
    16, 44.5, 71.7, 73, 82, 100), y = c(100, 89, 85, 60, 37, 32, 
    25, 0, 100, 89, 85, 60, 37, 32, 25, 0), animal = c("cat", "cat", 
    "cat", "cat", "cat", "cat", "cat", "cat", "dog", "dog", "dog", 
    "dog", "dog", "dog", "dog", "dog")), class = "data.frame", row.names = c(NA, 
    -16L), .Names = c("x", "y", "animal"))

library(ggplot2)
ggplot(dat) + theme_bw() +
    geom_area(aes(x=x, y=y, group=animal, fill=animal), alpha=.3)  

Oddly, if I subset just dog it plots correctly:

ggplot(subset(dat, animal=="dog")) + theme_bw() +
    geom_area(aes(x=x, y=y, group=animal, fill=animal), alpha=.3)  

Answer 1

I get the same error... Another possible work around is to use geom_ribbon.

ggplot(dat) + theme_bw() +
geom_ribbon(aes(x = x, ymin = 0, ymax = y, group = animal, fill = animal, position = "stack"), alpha=.3)  

This also gives the right plot

Answer 2

I'm not sure why you get that unexpected result. I got the same strange plot.

A possible workaround:

ggplot() +
  geom_area(data=dat[dat$animal=="dog",], aes(x=x, y=y, fill="red"), alpha=.3) +
  geom_area(data=dat[dat$animal=="cat",], aes(x=x, y=y, fill="blue"), alpha=.3) +
  scale_fill_discrete("Animal",breaks=c("red","blue"),labels=c("dog","cat")) +
  theme_bw()

which gives the desired result: