Opinion by
Mr. Justice Dean,
The parties were married on April 14, 1858, and cohabited together until August 8, 1893, when the libellant alleges that his wife wilfully and maliciously deserted him, and at the date of the commencement of this proceeding, August 9, 1895, had absented herself from Iris habitation for a period of more than two years. The wife, in her answer, denies the desertion, and avers that her husband deserted her. On November 6, 1895, the court appointed Albert B. Kelley, Esq., master. The decree for this appointment is not printed. We infer, however, from the designation “master,” the scope of his authority was to take testimony, find facts and suggest a decree, as he has done. This proceeding in divorce, being wholly statutory, the remedy prescribed by the statute must be pursued. When the jurisdiction of a court is conferred by statute, and the manner in which the jurisdiction shall be exercised is pointed out, courts are not at liberty to adopt the practice of other courts, either common law or statutory, to reach a decree. While the court may appoint an examiner to take testimony and report it, there is no authority under the act to appoint a master to find facts and suggest a decree. Long established practice throughout the commonwealth, resting often in rules of court, has settled the construction of the act, as authorizing the appointment of an examiner, but the whole legislation on the subject clearly *615intends, that the court shall not shift the duty of finding the facts to an appointee. Whether the marital contract shall be severed is the gravest of questions, not alone to the parties, but to the state, for the social structure rests upon it. It never was intended that judicial function should in any material degree be relinquished by conducting the proceedings before a master in his office, or that weighty judicial responsibility should be evaded by shifting it over to a member of the bar. We feel sure, a careful perusal of tbe statutes will convince any one of the correctness of these observations. The ability, learning and conscience of tbe court must be called into exercise before there can he a dissolution of this contract. While the witnesses may he examined, and their testimony reduced to Avriting by the examiner, the court must, before decree, be satisfied by its oaaui knoAvledge of the testimony that the averments of the libel have been piwed by full and competent evidence. It is not sufficient that they haAm been proved to the satisfaction of the examiner by Avitnesses -that tlie court neither saw nor heard. In this case, the master reports, that the respondent Avilfully and maliciously deserted libellant and absented herself from his habitation for a period of more than two years, and that libellant is entitled to a decree of diA’orce from the bonds of matrimony. On this the court granted a rule to sIioav cause why such decree should not he entered; this rule Avas made absolute in less than thirty days thereafter, and the entry made, “Divorce decreed.” We presume the court examined and considered the testimony, although this record does not show it. The opinion of this Court as to the rule that should control has been repeatedly expressed. In Angier v. Angier, 63 Pa. 450, this language is used: “Courts ought never to sever the marriage contract but where the application is made in sincerity and truth, for the causes set forth, and no other, and fully sustained by the testimony.” To the same effect are Jones v. Jones, 66 Pa. 494, Edmond’s App., 57 Pa. 232, Richards v. Richards, 37 Pa. 225, Sowers’s App., 89 Pa. 173, and many others. And this Court has, ever since the passage of the act of 1815, held it incumbent on it, on appeal from a decree of divorce, except where there has been an issue and jury trial, to review the testimony, and adjudge whether it sustained the complaint of the libellant. It has not adopted, in such appeals, the rule gener*616ally applicable to proceedings before a master or an auditor, that a finding of fact will not be disturbed except for manifest error. In every case in which the appeal was from a decree not based on the finding of a jury from testimony produced and the finding had in open court, under the instructions of the judge as to the law, it has taken up, analyzed and reviewed the testimony, and in nearly every case, has embodied its views in an opinion filed. The last case is Van Dyke v. Van Dyke, 135 Pa. 459, where the opinion of the Court is expressed by our Brother McCollum, after a careful scrutiny of the testimony taken before the examiner, with the result that the decree was set aside, because the fact of wilful and malicious desertion was not sustained. Therefore, of whatever drudgery the court of original jurisdiction may relieve itself in this class of cases, by the appointment of an examiner, neither it nor we can escape the burden of a careful consideration of the evidence, to ascertain if it do, in very truth, establish the statutory grounds for a divorce. As is said in Richards v. Richards, supra, “Never ought divorces to be easily obtained, for marriage is the most sacred of human relations, and should never be dissolved without clear proof of imperious reasons.”
On examination of the testimony we cannot concur with the finding of the examiner that respondent was guilty of wilful and malicious desertion. As noticed, the parties were married in 1858; two children were born unto them, one of whom, Mary, survives; the husband was a man of some wealth; he owned a house on Forty-first street, Philadelphia, in which they lived until 1885, when they removed to a small farm owned by him at Roseglen, Montgomery county; here, the family lived until September 7, 1893. As he had sold the farm in the July preceding and must give possession to the purchaser, they were compelled to leave it on the day named. The mother and daughter went back to the house on Forty-first street in Philadelphia; the husband went to board with his sister in the same city. This was the beginning of the alleged wilful and malicious desertion. What brought it about? We take up libellant’s testimony, without a single reference to that of his wife and daughter, and find these facts from his own express admissions : The residence at Roseglen was about ten miles from the city, three quarters of a mile from the village, and tha *617same distance from the railroad station; the neighborhood was sparsely settled; the house was not supplied with the necessities and comforts of that from which they had removed; the husband, for seventeen years, had ceased marital relations with his wife; for that time they had occupied separate beds, if not separate rooms, and their social intercourse was of the most formal kind; for eight years they had not gone together to any place of entertainment. This sort of life, at that solitary place, was the condition from April 1, 1885, until the spring of 1892. Being a native of England, he desired to visit that country in that year, and wanted his wife and daughter to go with him. They consented, on a promise by him, that on their return he would provide a more agreeable home than that at Roseglen. When they came back in the autumn, he violated his promise ; made no effort to prepare another home, in fact, made preparations to live on in the same one. His own evidence shows that he considered this promise as one not made to be kept. There' had occurred no change in his circumstances; he was amply able to have kept it if he had so chosen. Such a violated promise was no justification for desertion on part of the wife. We only advert to it, because it was a source and aggravation of the domestic troubles which followed and continued until the day of separation. Naturally, the broken promise produced exasperation on part of the wife, and the consequence was unseemly quarrels between them. Her tongue was her only effective weapon in stating her wrongs and in asserting her rights, and she used it pretty freely. The house on Forty-first street was occupied by her sister, and she now commenced making very frequent visits there, so that from September, 1892, to September, 1893, according to a memorandum carefully kept by him, they numbered one hundred and eighty; they were generally for a day at a time, the alternate days being spent at Roseglen. We have no doubt that his inference is correct; that by this conduct she was putting a pressure on him to compel the fulfilment of his promise to change his residence. About the last of July, 1893, he made sale of the Roseglen property, possession to be given September 7 following, on which day the family left. She and the daughter, as heretofore noticed, going to the house on Forty-first street, and he to rooms in the house of his sister in the same city. *618If there was any semblance of desertion this was the commencement of it, for they did not thereafter live under the same roof. The frequent visits to the city, and the removal of part of the furniture with the intention by both to live apart, did not warrant the averment that the desertion commenced August 8. They lived in the same house, on about the same terms as before, and either or both, up to September 7, might have abandoned the intention previously formed to separate. Why didnot their former relations, chilling and uncomfortable as they were, continue ? He testifies that six months before the sale of Roseglen, exasperated by her sharp tongue and a threat to leave him, he engaged rooms for himself at his sister’s; he then suggested to her in a note dated December 31, 1893, that boarding will be nice for a year or two, and that each one will have a free foot, as she and Mary wish to have. In another letter, February 8, 1893, after narrating all he has to complain of in her conduct, he says, if this is to continue, they had better live apart, and that under the circumstances he considers it best that he shall go into lodgings, rather than be subject to indignities under the same roof with her. To promote such a separation and have no clashing, he says, the daughter can have the piano, and she, the wife, all the rest of the furniture, except some little for himself to furnish a room. Their relatioiis continued, nevertheless, the same up until the removal, when he hired the carter to haul away for his wife the furniture which, with his consent, she selected, and of which he at the time made a careful inventory. She and her daughter took up their abode in the Forty-first street" house, and he his with his sister. Taking his own testimony, and the whole of it, in the most favorable view to him, his wife was ill-tempered, sharp-tongued, shocked him by scepticism in religion, and made his life unhappy; to obtain relief he suggested a voluntary separation to which she assented. Then, for two years, there followed much correspondence and occasional personal interviews, but no reconciliation sufficiently complete to re-establish the marital relation. He suggested conditions, she attempted to exact others; his letters, if candid, indicate discomfort and unhappiness at the continued separation. In the one of November 28, 1894, he says in reference to her demand for some "settled provision for the future comfort : “ What we made in partnership together I want you to *619have tbe full enjoyment of, as is your right.” By this ho refers to their early frugality and self-sacrifice in accumulating their fortune. He then unburdens himself thus: “ I feel it all the time, you have been kind to me, gentle and forbearing, no one knows better than myself.” But before filing his libel he, on April 3, 1895, by advice of counsel, as he says, gives her a formal written notification to return to his house, which he has purchased, and will soon be in possession of, and further, gives her one week to decide, with this supplementary warning: “ You will have to satisfy me you are willing to return to home, husband and father, and life’s duties, by abandoning the causes of this dereliction of duties.” Whether the refusal to return to him as requested by such a letter, and such refusal persisted in for the statutory period, would warrant a decree of divorce for wilful and malicious desertion it is not necessary to discuss. Assuming it to have been made in good faith, by the advice of counsel, on its face it would tend to repel acceptance rather than encourage it. But, however this may be, such refusal cannot constitute the wilful and malicious desertion of two years before averred in this libel, and therefore is not material.
In all we have said we have not noticed the testimony of respondent, which flatly denies wilful and malicious desertion, or any wilful and malicious refusal to return. On libellant’s own testimony, it was a separation suggested and encouraged by him. A separation, judging from the correspondence, after-wards repented by both; then, because of human frailties, a failure to be reconciled, although they had lived together for thirty-five years, made a comfortable fortune and reared one child.
As we have held over and over again, a separation is not, on the part of either, wilful and malicious desertion. In Ingersoll v. Ingersoll, 49 Pa. 249, it is decided, “ Separation is not desertion. Desertion is an actual abandonment of matrimonial cohabitation, with an intent to desert, wilfully and maliciously persisted in, without cause, for two years. The guilty intent is manifested when, without cause or consent, either party withdraws from the residence of the other.” And in Graham v. Graham, 153 Pa. 450, it was held that where the evidence showed that the wife gave her husband to understand that she did not want his society, and in consequence he lived apart *620from her five or six squares distant, this did not constitute wilful and malicious desertion on his part. And the same principle is applied hi many other cases.
As the evidence here wholly fails to sustain the averment of wilful and malicious desertion, the decree is reversed and libel dismissed at costs of libellant.