$q$-Matroids are defined on complemented modular support lattices. Minors of length 2 are of four types as in a "classical" matroid. Tutte polynomials $\tau(x,y)$ of matroids are calculated either by recursion over deletion/contraction of single elements, by an enumeration of bases with respect to internal/external activities, or by substitution $x \to (x-1),\; y \to (y-1)$ in their rank generating functions $\rho(x,y)$. The $q$-analogue of the passage from a Tutte polynomial to its corresponding RGF is straight-forward, but the analogue of the reverse process $x \to (x-1),\; y \to (y-1)$ is more delicate. For matroids $M(S)$ on a set $S$, and relative to any linear order on the points, the concept of internal/external activity of a point relative to a basis gives rise to a partition of the underlying Boolean algebra $B(S)$ into a set of "prime-free" (or "structureless") minors, such minors being direct sums of loops and isthmi (coloops), with one such prime-free minor for each basis. What usually goes unnoticed is that each prime-free minor has a unique clopen flat. The latter property carries over to $q$-matroids, but each prime-free minor will contain many bases. So internal and external activity in $q$-matroids must be defined not for points relative to bases, but rather for coverings in the underlying complemented modular lattice. Following lattice paths from arbitrary subspaces $A$ along active coverings (downward for internally active, upward for externally active) will lead to the unique clopen subspace in the prime-free minor containing the subspace $A$. There are a number of interesting questions concerning $q$-matroids that remain unsolved.