Full Idea
A famous example of Thomas Reid: day regularly follows night, and night regularly follows day. There is therefore a constant conjunction between night and day. But day does not cause night, nor does night cause day.
Gist of Idea
Day and night are constantly conjoined, but they don't cause one another
Source
report of Thomas Reid (Essays on Active Powers 1: Active power [1788]) by Tim Crane - Causation 1.2.2
Book Reference
'Philosophy: a Guide Through the Subject', ed/tr. Grayling,A.C. [OUP 1995], p.186
A Reaction
Not a fatal objection to Hume, of course, because in the complex real world there are huge numbers of nested constant conjunctions. Night and the rotation of the Earth are conjoined. But how do you tell which constant conjunctions are causal?