Indeed, Mr. Hume makes a great mistake in supposing it necessary to demonstrate, in every particular instance, what particular Effect must necessarily flow from its object, in order to gain the idea of necessary Connexion. The how and the why have nothing to do with the general reasoning affecting the general proposition; for "whether like Causes shall produce like Effects" is not a question exactly the same as whether " such particular causes shall have such particular effects? which Mr. Hume seems to consider as precisely of the same import; whereas one is a general question, which however answered, in the affirmative or negative, would apply to particulars.
Lady Mary Shepherd, An Essay upon the Relation of Cause and Effect, p. 59. This is one of Shepherd's most important criticisms of Hume's account of causation. The idea is that Hume conflates two different questions:
(1) Is it necessary that like causes have like effects?
(2) Is it necessary that this particular cause have this particular effect?
And her point is that, contrary to the way Hume seems to argue when discussing sensible qualities, in order to know that like causes have like effects, you don't need already to know what those effects would be in a particular case -- you can prove the general proposition, and then use it to investigate the particular case. Likewise, even if we knew the particular case, and could see precisely why this cause must have this effect, that would still leave open the general question of whether like causes have like effects. Hume isn't quite consistent: at one time he seems to talk about how we know that, supposing the causes similar, the effects must be similar, while at another he seems to talk about how we know that the causes themselves are similar in the first place. For instance, Hume says (ECHU, Section IV, Part II):
From causes which appear similar we expect similar effects. This is the sum of all our experimental conclusions. Now it seems evident that, if this conclusion were formed by reason, it would be as perfect at first, and upon one instance, as after ever so long a course of experience. But the case is far otherwise. Nothing so like as eggs; yet no one, on account of this appearing similarity, expects the same taste and relish in all of them. It is only after a long course of uniform experiments in any kind, that we attain a firm reliance and security with regard to a particular event.
But the problem here is that what leads us to the conclusion about eggs has nothing to do with the general principle that similar causes have similar effects: it's the fact that in the case of eggs it is hard to tell what counts as a similar cause. Even if the general claim is formed by reason, it can't be applied until we have identified the right causes and recognized them as similar. And this, Shepherd will argue, is a matter of reasoning on the basis of experiment and trial, an entirely different sort of thing than the sort of reasoning that undergirds the general claim.