After a brief introduction to Kuipers' views on explanations of laws we argue that micro-explanations of laws can have two formats: they work either by aggregation and transformation (as Kuipers suggests) or by means of function ascriptions (Kuipers neglects this possibility). We compare both types from an epistemic point of view (which information is needed to construct the explanation?) and from a means-end perspective (do both types serve the same purposes? are they equally good?).