> library(MASS)
> Iris <- data.frame(rbind(iris3[,,1], iris3[,,2], iris3[,,3]),
+                         Sp = rep(c("s","c","v"), rep(50,3)))
>      train <- sample(1:150, 75)
>      table(Iris$Sp[train])

c  s  v
26 21 28
>      ## your answer may differ
>      ##  c  s  v
>      ## 22 23 30
>      z <- lda(Sp ~ ., Iris, prior = c(1,1,1)/3, subset = train)
>      predict(z, Iris[-train, ])$class
[1] s s s s s s s s s s s s s s s s s s s s s s s s s s s s s c c c c c c c c c
[39] c c c c c c c v c c c c c c c v v v v v v v v v v v v v v v v v v v v v v
Levels: c s v
> str(predict(z, Iris[-train, ]))
List of 3
$ class    : Factor w/ 3 levels "c","s","v": 2 2 2 2 2 2 2 2 2 2 ...
$ posterior: num [1:75, 1:3] 1.22e-30 2.54e-24 7.86e-27 1.89e-30 1.48e-20 ...
  ..- attr(*, "dimnames")=List of 2
  .. ..$ :8322456 [1:75] "1" "2" "3" "6" ...
  .. ..$ :8322456 [1:3] "c" "s" "v"
$ x        : num [1:75, 1:2] -9.5 -8.21 -8.71 -9.37 -7.43 ...
  ..- attr(*, "dimnames")=List of 2
  .. ..$ :8322456 [1:75] "1" "2" "3" "6" ...
  .. ..$ :8322456 [1:2] "LD1" "LD2"
> predict(z, Iris[-train, ])$posterior
               c            s            v
1   1.218515e-30 1.000000e+00 2.370400e-54
2   2.536374e-24 1.000000e+00 3.220490e-46
3   7.862050e-27 1.000000e+00 2.130294e-49
6   1.894194e-30 1.000000e+00 2.190989e-53
9   1.477401e-20 1.000000e+00 2.779259e-41
10  1.301205e-24 1.000000e+00 6.038825e-47
11  7.346815e-33 1.000000e+00 3.016359e-57
13  1.585901e-24 1.000000e+00 7.917351e-47
14  2.554392e-26 1.000000e+00 4.718289e-49
15  1.097041e-42 1.000000e+00 5.797731e-70
16  2.553455e-40 1.000000e+00 4.043937e-66
17  1.824959e-36 1.000000e+00 3.784713e-61
18  6.817042e-30 1.000000e+00 5.053540e-53
19  8.796719e-32 1.000000e+00 1.557462e-55
20  1.581987e-31 1.000000e+00 4.243320e-55
21  1.029487e-26 1.000000e+00 2.384121e-49
22  9.842678e-30 1.000000e+00 1.973908e-52
23  5.529094e-35 1.000000e+00 7.422109e-60
............