This was a decision of the House of Lords (now the Supreme Court). Lord Pearson said, at page 301,
In an action for negligence the [claimant] must allege, and has the burden of proving, that the accident was caused by negligence on the part of the defendants.
That is the issue throughout the trial, and in giving judgment at the end of the trial the judge has to decide whether he is satisfied on a balance of probabilities that the accident was caused by negligence on the part of the defendants, and if he is not so satisfied the [claimant’s] action fails.
The formal burden of proof does not shift. But if in the course of the trial there is proved a set of facts which raises a prima facie inference that the accident was caused by negligence on the part of the defendants, the issue will be decided in the [claimant’s] favour unless the defendants by their evidence provide some answer which is adequate to displace the prima facie inference.
In this situation there is said to be an evidential burden of proof resting on the defendants. I have some doubts whether it is strictly correct to use the expression `burden of proof’ with this meaning, as there is a risk of it being confused with the formal burden of proof, but it is a familiar and convenient usage.
See other posts under the res ipsa tag: especially Ratcliffe.