The holistic answer to your question is contained here http://stats.stackexchange.com/a/23198/11523

To answer your specific question regarding creating contrast matrices

Once you have set options(contrasts=c("contr.sum","contr.poly"))

then you can use

 contrasts(x1)
##   [,1] [,2]
## 1    1    0
## 2    0    1
## 3   -1   -1
contrasts(x2)
##   [,1]
## 1    1
## 2   -1
contrasts(interaction(x1,x2))
##     [,1] [,2] [,3] [,4] [,5]
## 1.1    1    0    0    0    0
## 2.1    0    1    0    0    0
## 3.1    0    0    1    0    0
## 1.2    0    0    0    1    0
## 2.2    0    0    0    0    1
## 3.2   -1   -1   -1   -1   -1

to create valid contrast matrices and to recreate your tables by hand..

Read the answers (an links) from the linked questions to decide whether type III SS are useful or not.

To avoid the tedious hand-calculations, use the following (from http://stats.stackexchange.com/a/23198/11523)

model <- lm(y ~ pre + x1*x2)

drop1(.model, .~., test = 'F')
Single term deletions

Model:
y ~ pre + x1 * x2
       Df Sum of Sq    RSS     AIC F value  Pr(>F)  
<none>              2.8750 -3.1462                  
pre     1    3.1250 6.0000  3.6822  5.4348 0.06711 .
x1      2    4.5806 7.4556  4.2888  3.9832 0.09234 .
x2      1    3.0179 5.8929  3.4660  5.2484 0.07057 .
x1:x2   2    1.2500 4.1250 -2.8141  1.0870 0.40554  

The holistic answer to your question is contained here https://stats.stackexchange.com/a/23198/11523

To answer your specific question regarding creating contrast matrices

Once you have set options(contrasts=c("contr.sum","contr.poly"))

then you can use

 contrasts(x1)
##   [,1] [,2]
## 1    1    0
## 2    0    1
## 3   -1   -1
contrasts(x2)
##   [,1]
## 1    1
## 2   -1
contrasts(interaction(x1,x2))
##     [,1] [,2] [,3] [,4] [,5]
## 1.1    1    0    0    0    0
## 2.1    0    1    0    0    0
## 3.1    0    0    1    0    0
## 1.2    0    0    0    1    0
## 2.2    0    0    0    0    1
## 3.2   -1   -1   -1   -1   -1

to create valid contrast matrices and to recreate your tables by hand..

Read the answers (an links) from the linked questions to decide whether type III SS are useful or not.

To avoid the tedious hand-calculations, use the following (from https://stats.stackexchange.com/a/23198/11523)

model <- lm(y ~ pre + x1*x2)

drop1(.model, .~., test = 'F')
Single term deletions

Model:
y ~ pre + x1 * x2
       Df Sum of Sq    RSS     AIC F value  Pr(>F)  
<none>              2.8750 -3.1462                  
pre     1    3.1250 6.0000  3.6822  5.4348 0.06711 .
x1      2    4.5806 7.4556  4.2888  3.9832 0.09234 .
x2      1    3.0179 5.8929  3.4660  5.2484 0.07057 .
x1:x2   2    1.2500 4.1250 -2.8141  1.0870 0.40554  

The holistic answer to your question is contained here http://stats.stackexchange.com/a/23198/11523

To answer your specific question regarding creating contrast matrices

Once you have set options(contrasts=c("contr.sum","contr.poly"))

then you can use

 contrasts(x1)
##   [,1] [,2]
## 1    1    0
## 2    0    1
## 3   -1   -1
contrasts(x2)
##   [,1]
## 1    1
## 2   -1
contrasts(interaction(x1,x2))
##     [,1] [,2] [,3] [,4] [,5]
## 1.1    1    0    0    0    0
## 2.1    0    1    0    0    0
## 3.1    0    0    1    0    0
## 1.2    0    0    0    1    0
## 2.2    0    0    0    0    1
## 3.2   -1   -1   -1   -1   -1

to create valid contrast matricesto recreate your tables by hand.

Read the answers (an links) from the linked questions to decide whether type III SS are useful or not.

or you can use

model <- lm(y ~ pre + x1*x2)

drop1(.model, .~., test = 'F')
Single term deletions

Model:
y ~ pre + x1 * x2
       Df Sum of Sq    RSS     AIC F value  Pr(>F)  
<none>              2.8750 -3.1462                  
pre     1    3.1250 6.0000  3.6822  5.4348 0.06711 .
x1      2    4.5806 7.4556  4.2888  3.9832 0.09234 .
x2      1    3.0179 5.8929  3.4660  5.2484 0.07057 .
x1:x2   2    1.2500 4.1250 -2.8141  1.0870 0.40554  

to avoid all the by-hand calculations

The holistic answer to your question is contained here http://stats.stackexchange.com/a/23198/11523

To answer your specific question regarding creating contrast matrices

Once you have set options(contrasts=c("contr.sum","contr.poly"))

then you can use

 contrasts(x1)
##   [,1] [,2]
## 1    1    0
## 2    0    1
## 3   -1   -1
contrasts(x2)
##   [,1]
## 1    1
## 2   -1
contrasts(interaction(x1,x2))
##     [,1] [,2] [,3] [,4] [,5]
## 1.1    1    0    0    0    0
## 2.1    0    1    0    0    0
## 3.1    0    0    1    0    0
## 1.2    0    0    0    1    0
## 2.2    0    0    0    0    1
## 3.2   -1   -1   -1   -1   -1

to create valid contrast matrices and to recreate your tables by hand..

Read the answers (an links) from the linked questions to decide whether type III SS are useful or not.

To avoid the tedious hand-calculations, use the following (from http://stats.stackexchange.com/a/23198/11523)

model <- lm(y ~ pre + x1*x2)

drop1(.model, .~., test = 'F')
Single term deletions

Model:
y ~ pre + x1 * x2
       Df Sum of Sq    RSS     AIC F value  Pr(>F)  
<none>              2.8750 -3.1462                  
pre     1    3.1250 6.0000  3.6822  5.4348 0.06711 .
x1      2    4.5806 7.4556  4.2888  3.9832 0.09234 .
x2      1    3.0179 5.8929  3.4660  5.2484 0.07057 .
x1:x2   2    1.2500 4.1250 -2.8141  1.0870 0.40554