Guassian Mixture Classification: interpretating component x variable means matrix

The Guassian Mixture model output by mclust::Mclust() function has a $$parameters$$mean element which is a matrix with dimensions p x k, with p the number of variables of the original data and k the number of identified clusters.

What does this matrix represent?

• I also noted that these values are well correlated to the $P(p|k)$ retrievable in the \$z element of the mclust::Mclust() output – Bakaburg Oct 31 '18 at 17:37