Recently i analysed data from wheat trials conducted across locations. i calculated BLUEs and BLUPs for these genotypes and observed that some of the BLUEs or not the estimated values, infact they were actually simply the arithmetic means of the 2 reps data. i am looking for answer that what are the causes?

on the other hand for some locations (in individual analysis) BLUPs were not calculated, all the genotypes had one single value and that was grand mean of that location. can anybody explain this in a simple way, cause i am not a statistician :)