Analysis Using SAS 

To test equality of treatment effects, analysis using SAS may be performed using the following steps:

Data Input:
For performing analysis, input the data in the following format.  {Here call the block as BLK, treatment as TRT. It may, however, be noted that one can retain the same name or can code in any other fashion}.  

Prepare a SAS data file using 

Data Jorhat;   /*one can enter any other name for Data*/;
Input   BLK      TRT      Calories;
Cards;

1           2           3325060
1           4           2606200
1           5           3279420
1           6           2330180
2           1           2992900
2           2           3228180
2           5           3348780
2           7           2982000
3           1           2639980
3           4           2467800
3           6           2196580
3           7           2730780
4           3           2602410
4           5           3696340
4           6           2388060
4           7           2921790
5           1           3055180
5           3           2653680
5           4           2501060
5           5           3594320
6           2           3380420
6           3           2760690
6           4           2522100
6           7           2961270
7           1           2921420
7           2           3380420
7           3           2677400
7           6           2594420
;

/* To test whether there is any difference between treatments and blocks one can perform the analysis of variance of the data using the  following  steps.*/

    Proc glm;
class BLK TRT;
Model CALORIES = TRT BLK;
LSMEANS TRT/PDIFF;
Contrast 'treatment 1,2 vs treatment 3,4,5,6,7' Trt -5 -5 2 2 2 2 2;
Contrast 'within treatment 3,4,5,6,7' Trt 0 0 1 -1 0 0 0,
                                                       Trt 0 0 1 1 -2 0 0,
                                                       Trt 0 0 1 1 1 -3 0,
                                                       Trt 0 0 1 1 1 1 -4;
Run;

Here LSMEANS gives adjusted treatment means (least squares means) and pdiff performs all possible pair wise treatment  comparisons gives the exact probability level of significance in a v ´ v matrix. Hear v denotes the number of  treatments.
Type three sum of squares may be used for inferring on the quality of treatment and block effects.
We have written the contrasts in the program to answer the following.

i. Whether there is any significant difference between the average effect of 1^st two treatments (1,2) and the average effect of the remaining 5 treatments (3, 4, 5, 6, &7).

ii. Compare the treatment 3,4,5,6,7 among themselves.