delivered the opinion of the court.

• On the trial of this case the counsel for the defendant requested the court to charge the jury that to entitle the plaintiff’s to recover, the jury must be satisfied that the “scienter on the part of the defendant was proved;” which charge was excepted to by the counsel for the plaintiffs, from which we may infer that it was given by the court, although it is not so stated in the bill of exceptions.

It is insisted by the defendant’s counsel that the correctness of this charge must depend upon the form of action; whether it is ex contractu or ex delicto; that if it is in assumpsit on the warranty, the court erred; but if it is for deceit, the court was right. This position, though correct in part, is not so to the full extent. If it be an action of assumpsit predicated on a warranty, of course it was not necessary that the plaintiffs should have proved the scienter; and such proof was equally unnecessary if it be an i action of tort for deceit in a false warranty. Before I proceed to *291inquire into the form of the action, it may to as well to illustrate the truth of this position.

That an action for deceit can as well be sustained when there has been an express warranty, as where there has not, is a position deducible by reason and sustained by positive authority. In actions on the case purely for deceit in the sale of chattels, the deception is charged as resulting from the contract, made by false pretences and representations, amounting to a breach of the good faith by which every one is required to represent truly, the latent defects of the article he offers for sale, if known to him. And so it is when he gives a warranty. It is a matter of contract or engagement, which, if untrue, is more deceptive and fraudulent, because it inspires confidence by its professed fairness. In the one instance, the law has fixed a responsibility, and in the other, the party expressly engages to be responsible, implied by admitting a full knowledge of the qualities of the thing sold; and in cases of this description, it is unnecessary to prove the scienter, although the action be in tort, because of the parties undertaking to become responsible at all events; but the law does not impose such liability, in the absence of an express agreement, if the seller acted innocently; and it therefore requires proof that he knew of the defect.

In the case of Williamson v. Allison, 2 East, 446, the distinction is clearly taken. That was an action in tort for a breach of warranty, and the scienter was also laid in the declaration. For the defendant it was insisted that it was necessary that it should be proved, but the court held that it was unnecessary; that the breach of warranty was the gist of the action, and the scienter mere matter of aggravation, which might be stricken out, without affecting the declaration, and it was, therefore, considered as an immaterial averment. It was also held that the form of J action does not vary the proof. If, therefore, the gist of the present action be the breach of warranty the proof would be the same, whether it be in assumpsit or tort. In Selwyn’s Nisi Prius, 535, it is said that an action in tort may be sustained against any one who deceives by a false warranty, and the same doctrine is recognised hi Hallock v. Powell, 2 Caines’s Reports, *292216, and in the case of the Executors of Everston v. Miles, 6 Johns. Rep. 138. The books of forms furnish us, also, with precedents in this form of action for false warranty. The rule adopted by the English courts is, that actions of tort for false warranty are considered as arising ex contractu, and the party is held to the same rule of proof as though he had brought assumpsit. Weall v. King, 12 East, 152.

The declaration in this case is rather loosely drawn, but there can be no doubt but that it was designed to be in form ex de-licto. I am strengthened in this conclusion by having compared it with the precedents, furnished in tort for false warranty in 2 Chitty’s Pleading, 679, where the form will be found from which, I presume, this declaration was taken. The scienter is omitted by the author, and so it is in the first count in the declaration, and it is so untechnically laid in the second count, that it is difficult to tell whether it was intended by the pleader to perform that office or not. Adopting the rule held in Williamson v. Allison, it is clear that it might be stricken out without affecting the cause of action, and consequently it needed no proof. The false and fraudulent warranty is substantially charged in both counts, and forms, evidently, the gist of the action. The counsel for the defendant conceded in argument, that the action was in tort, but contended that it was, therefore, necessary that the scienter should have been proved. If we rely on the authorities cited, such proof was unnecessary, and the party did not by electing to sue in tort, change the nature of the proof. I have not been so fortunate as to find the authority in Hardin’s Reports, referred to by defendant’s counsel, and a different rule may be there adopted; but I should hesitate much before I could receive it, and reject the authority cited from 2 East.

The judgment must be reversed and a venire de novo awarded.