Hi Maria,
As described in the function documentation, each plot within a K represents a distinct "mode" in the space of possible solutions (see below). You can access the function documentation by typing in the R console:
> ?gl.plot.structure
For more information, you can read the references mentioned at the end of the function documentation.
Relevant information in the function documentation:
The Clumpak method identifies sets of highly similar runs among all the replicates of the same K. The method then separates the distinct groups of runs representing distinct modes in the space of possible solutions.
The CLUMPP method permutes the clusters output by independent runs of clustering programs such as structure, so that they match up as closely as possible.
This function averages the replicates within each mode identified by the Clumpak method.
Cheers,
Luis