delivered the opinion of the court.
The errors assigned in this cause, No. 17,366, are comprised in the two statements: First, that the order of dismissal on March 10, 1910, described in the preceding statement of facts was erroneous; and second, that it was void.
To the first of these propositions we give our assent; from the second we must withhold it.
We may pass over with a mere mention the question whether a writ of error should or can be properly sued out to reverse a void order. A “void” order seems necessarily to be a nullity which requires no reversal.
We prefer, however, since the decision of other causes connected with the one at bar depends on it, herein to state why we consider the order erroneous and why we do not consider it void. The contention that it is either, rests on the fact that on November 19, 1909, by consent of the parties, the order of reference to a Master described in the statement was made.
Counsel argue that an order of dismissal made after that order of reference, without a previous express vacation of the latter, was inconsistent therewith (which may be conceded), and void in consequence of such inconsistency (which is an entirely different proposition.)
The facts that the cause had been referred by the *536court, under an agreement of the parties, to a Master after it had been placed on the trial calendar of November 16, 1910, and that the hearing of evidence by him from time to time had gone on during several sessions before the order of March 10, 1910, was entered, and that neither party appeared on that last named date, when the cause appeared in the Law Bulletin on the trial calendar, or learned that any such order had been entered until long after many other sessions before the Master had been held, make apparent to us that the order of March 10, 1910, was entered by inadvertence. It is impossible for us to suppose that in the actual absence of both parties, the Chancellor,. had his attention been directed in any manner to the exact situation of the case, would have dismissed the bill for want of prosecution when he had less than four months before referred it to a Master to take evidence and state conclusions, and the complainant had not closed its proof, nor any complaint of want of expedition had been made by the defendant. Had the docket of the court been before him, showing the situation of each case as it was called on the trial calendar, as is practicable and usual for example in country districts, where the business of the courts is small, no such order as that of March 10, 1911, would have been entered. That it was so entered was undoubtedly because no notation had been made, on the trial calendar from which the court was calling the cases, of the order of November 19, 1910, referring the case to the Master.
Section 21 of the Practice Act of 1907 does not affect the matter; neither the provision that “all causes shall be tried or otherwise disposed of in the order they are placed on the docket unless the court for good and sufficient cause shall otherwise direct,” nor the further one that “no suit or action or proceeding at law or equity shall be dismissed for want of prosecution at any time except when such cause shall be ac*537tually reached for trial in its order as set for trial or upon the short cause or daily trial calendar of the court,” can in reason be held to mean that a judge finding, on reaching a chancery case on the docket, that it has been referred to a Master, may not pass it without action and proceed to dispose of cases beyond it.
We are not obliged to pass on the question whether when an order of this kind is plainly entered by inadvertence, as we think this one of dismissal was, it is ipso facto erroneous. For certainly the order cannot be less erroneous because entered by inadvertence than it would be-had it been entered with full knowledge and intention and in the exercise of the discretion of the court. If this order was not entered by inadvertent mistake, then it was entered as an exercise of the discretion of the court, and, in our opinion, if this was the case, was an abuse of discretion. It is needless to discuss this point, as the facts speak plainly for themselves.
The proposition that the order was erroneous under the circumstances we think the reasoning of many cases cited by the appellant, such as Cook v. Gwyn, 3 Atkins, 689; Bailey v. District of Columbia, 4 App. Cas. D. C. 358; Dexter v. Young, 40 N. H. 130; Tyson v. Robinson, 25 N. C. 333; Wilby v. Durgen, 118 Mass. 64, and others, fully sustains. Cases dependent on the law concerning references to arbitrators, however, although the reasoning may be applicable, are hardly authoritative on the question of reference to a Master. We think, moreover, that these cases are far from sustaining the position that the order was void.
When a court has taken, through sufficient process, jurisdiction of the persons of litigants in a subject-matter of which it also has jurisdiction, something beyond a reference to a Master in Chancery “to take evidence and report his findings and conclusions of law to this court” is necessary to oust it of jurisdiction. And if when it made the order of March 10, *5381910, the court had jurisdiction to make it, it is not void nor a nullity, however erroneous it may be.
A Master in Chancery, as the Supreme Court through Mr. Justice Scholfield in Schuchardt v. People, 99 Ill. 504, said, quoting approvingly from Bouvier’s Law Dictionary, was originally a clerk for a chancellor, and became afterwards an assistant to the chancellor, and has been invested with other powers by local regulations. His function, as the same court said in Ennesser v. Hudek, 169 Ill. 494, is primarily to perform clerical and ministerial duties in the progress of a case.
The court had power on March 10, 1910, to do exactly as it did do with the cause, namely, dismiss it at complainant’s costs, but its action was a misuse of that power and was erroneous.
We think that none of the cases cited by the plaintiff in error to show that the order was erroneous will be found on close examination to ignore this distinction or negative its existence.
When in Gregory v. Spencer, 11 Beav. 143, Lord Langdale said, “I can make no order,” it is clear that he meant “I can properly make no order,” and so we hold in this case. But we can go no further.
What Judge Cowen said in Gould v. Root, 4 Hill (N. Y.) 554 (as cited by plaintiff in error), if slightly changed would fit this case and the order under discussion :
“I am not aware of any principle which authorizes a party to treat an order as a nonentity merely because a commissioner is forbidden to grant it or a party is forbidden to apply for it,” or a ju,dge should not have granted it.
“It may be said of every order improvidently granted that the party and officer” or judge “have done what the law has forbidden. To say that it was fraudulently and collusively obtained is no more. But to allow as a consequence that it may therefore be disregarded would be letting in a principle under which every judicial act might be questioned collaterally.
*539The remedy is by direct proceeding, which in the case of orders is revocation,” if ashed while the judge retains jurisdiction, “appeal” or writ of error “or motion to supersede” if ashed in time.
The italics represent that which we have added to the quotation to make it exactly fit the case at bar, and with those additions it is as true as without them. As the same judge said in Starr v. Frances, 22 Wend. (N. Y.) 633, of the order he was then discussing, we may say of the order of dismissal in this case: “It is a judicial act done in the course of the cause, opén to a rehearing and appeal to this court, and we are not aware of any case in which such an act has been holden void even for fraud.”
For the reason, however, that it was erroneous it is reversed and the cause remanded to the circuit court for such othe$ and further proceedings therein as to justice shall appertain.
Reversed and remanded.