High Court of Australia Barwick C.J. Gibbs, Stephen, Mason and Jacobs JJ. Russell and Russell [1976] HCA 23 ORDER Russell v. Russell The questions referred to the Full Court are answered as follows: 1. Is (i) section 97 in so far as it deals with courts other than the Family Court (ii) section 9(4) (iii) section 39 (iv) section 64 (v) section 74 (vi) section 78 of the Family Law Act 1975 a Law of the Commonwealth validly enacted pursuant to the powers conferred by pars (xxi.), (xxii.) and (xxxix.) of s. 51, s. 77 (iii.) or any other provision of the Constitution? Answer: (i) No, in relation to sub-s. (1). Yes, in relation to sub-s. (4). (ii) Yes, but only to the same extent as s. 39. (iii) Yes, s. 39 is a valid law of the Commonwealth in so far as the jurisdiction conferred relates to the following paragraphs of the definition of matrimonial cause contained in s. 4: (1) Paragraphs (a), (b) and (d). (2) Paragraphs (c) (i) and (e) but only in so far as they relate to proceedings between the parties to a marriage. (3) Paragraph (c) (ii) but only to the extent to which it relates to proceedings for ancillary relief in relation to proceedings of the kinds referred to in pars (a) and (b) of the definition. (4) Paragraph (c) (iii) but only in so far as it relates to proceedings between the parties to a marriage with respect to the custody, guardianship or maintenance of, or access to, the natural and adopted children of the parties to the marriage. (5) Paragraph (f) but only to such extent as is consequential upon the validity of s. 39 in relation to pars (a), (b), (c), (d), and (e) of the definition of "matrimonial cause". (iv) Yes, but only to the extent to which it relates to proceedings between the parties to a marriage with respect to the natural and adopted children of the parties to the marriage. (v) Yes, in so far as it relates to proceedings between the parties to a marriage or between the parties to a marriage, or one of them, and the natural or adopted child of that marriage. (vi) Yes, but only in so far as such proceedings are brought to obtain relief as an ancillary to proceedings of the kinds referred to in pars (a) and (b) of the definition of "matrimonial causes". 2. Is s. 123 of the said Act to the extent that it authorizes the making of regulations for or in relation to the practice and procedure to be followed in courts other than the Family Court exercising jurisdiction under the Family Law Act 1975 beyond the power of the Parliament of the Commonwealth to any and if so what extent? Answer: Not necessary to answer. 3. In so far as s. 97 of the said Act is a valid law of the Commonwealth does non-compliance with the provisions of (a) sub-section (1) (b) sub-section (3) (c) sub-section (4) thereof by a court (other than the Family Court) exercising or purporting to exercise jurisdiction under the Act affect the validity of any decree or order of such Court made in exercise or purported exercise of such jurisdiction? Answer: (a) Does not arise. (b) Not necessary to answer. (c) No. Remit case to Stephen J. for remission to the Supreme Court of Victoria pursuant to the request of counsel for the petitioner and to s. 45 of the Judiciary Act 1903 as amended. Farrelly v. Farrelly Remit case to the Supreme Court of South Australia to be heard and determined in accordance with the provisions of the Family Law Act 1975, pursuant to the request of counsel for the petitioner and to s. 45 of the Judiciary Act 1903, as amended. Cur. adv. vult. The following written judgments were delivered:— May 11 Barwick C.J. These two matters were removed into this Court by virtue of s. 40A of the Judiciary Act 1903-1973 Cth ("the Judiciary Act"). The first is a case brought in the Supreme Court of Victoria in which dissolution of marriage was sought by a wife on the grounds of her husband's adultery and of desertion, and also ancillary orders for maintenance of the petitioner, custody and maintenance of and for the five children of the marriage and for an order for settlement of property. There was a cross-petition seeking dissolution on the ground of desertion and dismissal of the claim for maintenance, custody and settlement of property. The second is a case brought in the Supreme Court of South Australia under the Guardianship of Infants Act, 1940 S.A.. The applicant in the case had been granted by a magistrate the custody of a child of her marriage to the respondent in the case. However, upon the commencement of the proceedings under the Guardianship of Infants Act, the Supreme Court made an order ex parte granting custody of the child to the applicant. Thereafter the respondent applied for the discharge of that order and for an order that he be given the custody of the child. Issues were raised in each case involving questions inter se within the meaning of the Constitution and the Judiciary Act. On removal, the matters were heard together by the Court. The first matters raised the validity of s. 97 of the Family Law Act 1975 Cth ("the Act"), and both raise the question of the validity of s. 39 in relation to so much of the definition of "matrimonial causes" as appears in pars (c), (d), (e) and (f) of s. 4(1) of the Act. The relevant parts of s. 97 are in the following terms: (1) Subject to sub-s. (2) and to the regulations, all proceedings in the Family Court, or in another court when exercising jurisdiction under this Act, shall be heard in closed court. (2) Subject to the regulations, relatives or friends of either party, marriage counsellors, welfare officers and legal practitioners may be present in court unless in a particular case the court otherwise orders. (3) In proceedings under this Act, the court shall proceed without undue formality and shall endeavour to ensure that the proceedings are not protracted. (4) Neither the Judge hearing proceedings under this Act nor counsel shall robe. The relevant parts of s. 39 provide: (1) Subject to this Part, a person may institute a matrimonial cause under this Act— (a) in the Family Court; or (b) in the Supreme Court of a State or a Territory. (2) Subject to this Part, a person may institute a matrimonial cause under this Act, not being proceedings for principal relief, in a court of summary jurisdiction of a State or Territory. (5) Subject to this Part, the Supreme Court of each State is invested with federal jurisdiction, and jurisdiction is conferred on the Family Court and on the Supreme Court of each Territory, to hear and determine— (a) matrimonial causes instituted under this Act: (b) matrimonial causes continued in accordance with s. 9; and (c) proceedings instituted under regulations made for the purposes of s. 106, 109, 110 or 111. (6) Subject to this Part, the courts of summary jurisdiction of each State are invested with federal jurisdiction, and jurisdiction is conferred on the courts of summary jurisdiction of each Territory, to hear and determine— (a) matrimonial causes, not being proceedings for principal relief, instituted under this Act; (b) matrimonial causes, not being proceedings for principal relief, continued in accordance with s. 9; and (c) proceedings instituted under regulations made for the purposes of s. 106, 109, 110 or 111. The definition of matrimonial cause in s. 4(1) of the Act is as follows: "matrimonial cause" means— (a) proceedings between the parties to a marriage for a decree of— (i) dissolution of marriage; or (ii) nullity of marriage; (b) proceedings for a declaration as to the validity of a marriage or of the dissolution or annulment of a marriage by decree or otherwise; (c) proceedings with respect to— (i) the maintenance of one of the parties to a marriage; (ii) the property of the parties to a marriage or of either of them; or (iii) the custody, guardianship or maintenance of, or access to, a child of a marriage; (d) proceedings between the parties to a marriage for the approval by a court of a maintenance agreement or for the revocation of such an approval or for the registration of a maintenance agreement; (e) proceedings for an order or injunction in circumstances arising out of a marital relationship; or (f) any other proceedings (including proceedings with respect to the enforcement of a decree or the service of process) in relation to concurrent, pending or completed proceedings of a kind referred to in any of paragraphs (a) to (e), including proceedings of such a kind pending at, or completed before, the commencement of this Act. Section 8(1) provides: After the commencement of this Act— (a) proceedings by way of a matrimonial cause shall not be instituted except under this Act; and (b) proceedings by way of a matrimonial cause instituted before the commencement of this Act shall not be continued except in accordance with section 9. and s. 9(4) provides: Pending proceedings constituting a matrimonial cause, not being proceedings for principal relief, whether instituted under the repealed Act or under the law of a State or Territory, may be continued and shall be dealt with as if they were proceedings instituted under this Act. Upon the first mentioned matter being removed into this Court, my brother Stephen referred to the Full Court the following questions and directed the parties to file written submissions with respect to them: 1. Is (i) section 97 in so far as it deals with courts other than the Family Court (ii) section 9(4) (iii) section 39 (iv) section 64 (v) section 74 (vi) section 78 of the Family Law Act 1975 a law of the Commonwealth validly enacted pursuant to the powers conferred by pars (xxi.), (xxii.) and (xxxix.) of s. 51, s. 77 (iii.) or any other provision of the Constitution? 2. Is s. 123 of the said Act to the extent that it authorizes the making of regulations for or in relation to the practice and procedure to be followed in Courts other than the Family Court exercising jurisdiction under the Family Law Act 1975 beyond the power of the Parliament of the Commonwealth to any and if so what extent? 3. In so far as s. 97 of the said Act is a valid law of the Commonwealth does non-compliance with the provisions of (a) sub-section (1) (b) sub-section (3) (c) sub-section (4) thereof by a Court (other than the Family Court) exercising or purporting to exercise jurisdiction under the Act affect the validity of any decree or order of such Court made in exercise or purported exercise of such jurisdiction? There is no attack in these cases upon the Act as a whole. The attack is on the validity of specific sections. Although in the course of argument other sections were referred to and discussed, the Court is not called upon in these cases to pronounce upon the validity of those provisions. Any decision as to the validity of such provisions of the Act must await proceedings in which the relevant sections are challenged and a decision as to their validity becomes necessary for the resolution of the questions arising in such proceedings. Section 39, in so far as it purports to vest federal jurisdiction in State courts, is, in my opinion, severable from so much of the same section as purports to give jurisdiction to federal courts. Consequently, a decision that s. 39 or s. 97 is invalid as to certain aspects of the invested jurisdiction and as to some proceedings does not itself affect the operation of those sections in other respects. The constitutional power which would support investiture of federal jurisdiction in State courts is to be found in Ch. III of the Constitution. It is not to be found in ss. 51(xxi.) or 51(xxii.) or in any other paragraph of s. 51. Legislation made upon the subject matter of paragraphs of s. 51 may delineate a jurisdiction but its investiture in State courts must take place, in my opinion, by virtue of s. 77(iii.) found in Ch. III. The power there given is to invest State courts with federal jurisdiction. Federal jurisdiction is the power or authority to hear and determine the matters validly specified in the legislation. It is to State courts that this jurisdiction is to be given. Such courts are courts created, maintained and organized under State law. Section 97 purports, amongst other things, to require the State court judge not to robe himself as State judges may do and not to sit in open court as State judges necessarily do, subject to the exercise of certain statutory discretions to close their courts on specific occasions. The robing of the judge is not required by any State statutory law but is part of the traditional inheritance of the State court as a court in the common law tradition. Again, no statute requires the court to sit in a place open to the public but, none the less, it is of the essence of the State system of courts that, unless authorized by statute, the place where a State court sits to exercise its jurisdiction will be open to the public. Certain discretions are given to the State courts to exclude members of the public from the proceedings in certain circumstances. The very existence of such a statutory discretion is emphatic that in general the court during the hearing of a matter is to be open to the public. A question was raised in the course of the argument of these matters, as to whether, in the event that this Court thought that s. 97 of the Act was valid in relation to State courts invested with jurisdiction under the Act, the proceedings of a court which had not conformed to the provisions of that section might in consequence be void. But, in my opinion, neither the validity nor the regularity of what the court does would be affected by the wearing of or the dispensation with, judicial robes. If, without any discretion so to do in the circumstances, the court or judge sat in a place not open to the public, that course would be such an irregularity as might provide ground for an appeal in which the court's or the judge's orders might be set aside for error or perhaps, in appropriate circumstances, for prohibition or certiorari. But what the court or judge did in those circumstances would not for that reason be void though it might be voidable on appeal or by prohibition or certiorari. It seems to me that s. 97 does not form part of the process of investing jurisdiction in State courts. I am quite clear that the section, in so far as it purports to command the court in the matter of dress, either of the judge or of counsel, is beyond the power of the Parliament. It is, in my opinion, inherent in the concept of a court that it can decide for itself how its members will be attired, and what dress it will expect of those who are to appear before it as representatives of parties. This is not a matter which, in my opinion, goes to the nature of the jurisdiction of the court, nor to the manner of the exercise: it is not relevantly part of the rules of procedure for the hearing of a matter. In my opinion, the command attempted to be given by s. 97(4) can be disregarded. It is for the State courts to decide for themselves whether their judges will or will not robe, unless the State legislature should otherwise provide. The courts of the States, as I have said, are in general required, because of the nature of the courts themselves and of the functions they perform, to sit and exercise jurisdiction in a place open to the public. The Parliament, in my opinion, has no power or authority to command the court of the State to sit in a place to which the public is not admitted. Whether a court should be closed for the purpose of certain proceedings involves a question of policy: with that I am not concerned. Much may be said pro and con as to the wisdom of closing a courtroom to the public in any but the most demanding circumstances. But in relation to State courts, in my opinion, it is for the legislature of the State alone to determine that policy and to prescribe the occasion when jurisdiction may be exercised in a closed court. Section 97(1) purports to take from the State court any discretion to exclude the public in certain particular circumstances, a discretion which when it is available, in my opinion, inheres in the nature and organization and operation of the court. The exercise of such a discretion does not touch upon the nature or extent of the jurisdiction to hear and determine. The section of the Act purports to deny the court of the State any such discretion and to require that in all circumstances in which the invested jurisdiction under the Act is exercised, the court shall exercise it behind closed doors. This, in my opinion, does not relate to and is not comprised in, the investiture of federal jurisdiction in the State court: it is not part of the grant of authority to hear and determine specific matters. In my opinion, therefore, s. 97 is invalid, in so far as it purports to require the State court to be closed to the public, when invested jurisdiction under the Act is being exercised. As I have indicated, this conclusion does not affect the validity of the section in relation to federal courts created under s. 71 of the Constitution. I should add expressly, lest it should not be appreciated that I have implicitly said so, that the wearing of robes and the place and nature of the place where the court sits are not, in my opinion, in any relevant sense part of the rules of practice or procedure of the court in exercising its jurisdiction in the hearing and decision of a case. They relate to the court's own organization and not to the procedure to be adopted in determining a matter. I now turn to the question which is of greater consequence, namely, whether, under either or both of the pars (xxi.) and (xxii.) of s. 51, the Parliament may create a jurisdiction to grant maintenance for a married person or a child of the marriage as defined in the Act, to grant custody of such a child and to determine the rights of the property of a married person, again as defined (see s. 4(2)), where there are or have been no proceedings for dissolution or nullity of marriage by or against the married person to which the proceedings for maintenance, custody or settlement of property can be regarded as ancillary. Section 39 purports to give the federal courts, and to invest State courts with, power to hear and determine matrimonial causes as defined in the paragraphs of the definition of "matrimonial cause" which I have set out. No attack has been made in these cases upon the validity of s. 39 in so far as matrimonial causes within pars (a) and (b) of the definition are concerned: but it is said that, upon their true construction, the pars (c), (d), (e) and (f) are unrelated to present or past proceedings for divorce or nullity of marriage: that is to say, to any proceedings as defined in pars (a) and (b) of the definition. In consequence, it is submitted that s. 39 does not validly create or invest a jurisdiction in any court to entertain such proceedings. There can be no doubt that, in terms, those paragraphs of the definition are unrelated to any proceedings for principal relief. They are not confined in terms to what is commonly called "ancillary relief" where the proceedings for dissolution or annulment are regarded as "principal relief". They thus confine proceedings for forms of relief appropriate as ancillary to principal relief but which in terms of the definition may themselves be the only relief which may be granted. The question, therefore, is whether a law creating a jurisdiction to make orders for maintenance, custody or settlement of property where the application therefor is unrelated to any proceedings for divorce or nullity is a law with respect to marriage: clearly, such a law cannot fall within the powers granted by par. (xxii.) of s. 51. The submission of the Solicitor-General for the Commonwealth would affirm that such proceedings fall within the power granted by par. (xxi.). The parties to these proceedings and the Solicitors-General for the States of Victoria and Western Australia submit to the contrary. Whilst each topic referred to in s. 51 is an independent subject matter and some overlapping is possible, cognate topics and the terms in which they are expressed cannot be and never have been ignored in deciding the content and ambit of a topic described in s. 51. In the present instance, the presence and the terms of par. (xxii.) may be related to the content and ambit of par. (xxi.) and may be regarded, as I would regard them, as limiting that content and ambit, particularly with respect to proceedings in relation to parental rights, custody and guardianship of infants. As I shall later point out, par. (xxii.) covers "matrimonial causes" as a specific head of power. In specifying that topic, the Constitution expressly provides the extent to which parental rights, custody and guardianship of infants fall within the area of matrimonial causes. It seems to me to be implicit in the topic of "divorce and matrimonial causes" that proceedings incidental to a proceeding for divorce or nullity of marriage are matrimonial causes within the subject matter of par. (xxii.) and that it is not necessary to resort to s. 51 (xxxix.) in order to find power to create a jurisdiction with respect to such proceedings. These undoubted consequences of par. (xxii.) are pertinent, in my opinion, to any consideration of the content or ambit of the power granted by par. (xxi.). Now, it is quite true that the topic "marriage" does not confine the Parliament to the making of laws with respect to the celebration of marriage. In my opinion the power of the Parliament extends to attaching consequences to the act of marriage, both for the spouses and through their parent or parents, for the children of one or both of them. It may also extend to the creation of mutual rights and obligations to flow from the act of marriage. I would respectfully agree with the ground upon which Sir Frank Kitto placed the legitimation provisions of the Marriage Act 1961 Cth (ss. 89-93) when, in Attorney-General (Vict.) v. The Commonwealth, his Honour said [16] : But, however this may seem in regard to legitimating steps of other kinds, at least it should be conceded, I think, where a marriage between a child's parents is made to spell the legitimation of the child. For it is of the essence of marriage, from a legal point of view that it produces, or provides a pre-requisite for, the legal recognition of family relationships; and what a law does which provides for legitimation by marriage is simply to add to the legal significance of marriage in this very matter of legal relationships. It is not as if, under such law, the change of status as between child and parents depended upon the intention of agreement of the parents when marrying one another. If it did, marriage, though referred to in the enactment, could hardly be described as a subject in respect of which the law was made. But a law which makes the legitimation of a child—perhaps one might more appropriately say the legitimation of the parents as such—an inevitable legal consequence of the intermarrying of the parents seems to me to be a law directly and squarely upon the subject of what marriage amounts to in law, and therefore upon the subject of marriage. Whether a law operating by reference to the married status, a Married Women's Property Act for example, is also a law upon marriage is a question of a different kind, and I say nothing about it. Here we are concerned only with a law dealing with the legal nature of marrying, a law joining with other laws to fix the bounds of the legal changes which marrying is to bring about. Though other Justices who participated in the decision of that case expressed themselves in different terms, I do not understand any of them to have gone beyond what Sir Frank Kitto had to say. If, contrary to my own view, it is thought that they did, then, in my opinion, their remarks should be regarded as obiter because all that was necessary for decision in that case was whether Parliament could attach to the act of marriage a consequence which affected the spouses and their child or children. 1. (1962) 107 C.L.R. 529, at p. 554. In my opinion, however, it is one thing to specify the consequences of the act of marriage, including a specification of the mutual rights and duties of the spouses, both towards each other and towards their children: it is quite another thing to erect a jurisdiction to enforce those rights and duties. Though in relation to some topics of legislation powers of adjudication and enforcement may be seen as no more than incidental to the topic or to the execution of the law made under it, it seems to me that the creation of such a jurisdiction, as that with which I am dealing, is not incidental within the meaning of par. (xxxix.) of s. 51 to the declaration of the consequences of marriage: nor is it wrapt up as an incident of and contained within the constitutional topic of "marriage" itself. It is, indeed, quite a disparate matter to determine how and by whom the particular consequences of the existence or exercise or non-performance of the rights or duties derived from the marriage may be determined and enforced. Hence the need for a specific topic of divorce and a specific topic of matrimonial causes, expressed in the terms of par. (xxii.) of s. 51. The limitations which, in my opinion, exist within the topic of marriage are pointed up by the inclusion of the proceedings covered by par. (c) (ii) of s. 4(1) within the scope of the jurisdiction created through s. 39. Though a system of communal property between spouses might possibly be erected as a consequence of the act of marriage—I have no need to decide nor to concede that such a course is possible—it could not properly be said, in my opinion, that the creation of jurisdiction to settle disputes as to such property owned or claimed by one or other of the spouses, was within the subject matter "marriage". They may well fall within "matrimonial causes" in par. (xxii.). But, in the Act, it is sought to create such a jurisdiction as to property in which no interest is derived from the act of marriage. It is enough for the purpose of s. 78 of the Act that the parties to the proceedings are or have been married (see s. 4(2)). In my opinion, there is no legislative power under par. (xxi.) to create a jurisdiction to determine such a dispute: nor, if it matters, would such a dispute give rise to a matrimonial cause within par. (xxii.). Further, in my opinion, there is no power within par. (xxi.) to create a jurisdiction to hear a dispute between spouses as to property the right or claim to which arises out of the marriage. Thus, for the reason that the ambit of the constitutional topic "marriage" does not extend to the creation of a jurisdiction to deal with a lawful obligation arising out of the act of marriage, in my opinion, the creation of a jurisdiction to entertain and determine proceedings for maintenance, custody and settlement of property is not authorized by par. (xxi.). The matter is even plainer, to my mind, in the case of proceedings unrelated to proceedings for divorce or nullity of marriage. These are not matrimonial causes within the ambit of par. (xxii.) and, in any case, may not, in my opinion, be authorized under par. (xxi.) properly understood. The matter may also be approached by considering as a separate matter the effect of the presence and the terms of par. (xxii.) upon the proper construction of the topic of par. (xxi.). The express inclusion of proceedings in respect of parental rights, custody and guardianship of infants within the topic of matrimonial causes in par. (xxii.), and the express limitation of those proceedings as matrimonial causes to ancillary proceedings seem eloquent to my mind, first, that only ancillary proceedings for maintenance, etc. are matrimonial causes under the Constitution and that the enforcement of parental rights, custody etc. does not fall within the subject matter of "marriage". To read par. (xxi.) as authorizing the creation of jurisdiction to entertain matrimonial causes must, in my opinion, treat par. (xxii.) as otiose and unnecessary. For, if the power given by par. (xxi.) is large enough to include the erection of a jurisdiction to hear and determine matrimonial causes, in the sense of disputes between husband and wife as to their mutual relationships, rights and obligations stemming from the fact of their marriage, and presumably between child and parent, then no significance remains for so much of par. (xxii.) as gives power with respect to matrimonial causes: and, indeed, also so far as it includes divorce. The suggested view would necessarily stretch to all matrimonial causes. Such a result, in my opinion, would be completely opposed to the principles of constitutional construction. Paragraph (xxi.) must be read in the presence of par. (xxii.) and not, as it were, in disregard of it. There would be no room, in my opinion, for treating matrimonial causes so far as they did not fall within par. (xxii.) as comprehended in par. (xxi.). In my opinion, had par. (xxii.) not appeared in s. 51, the legislative power granted by par. (xxi.) would not have extended to include any matrimonial cause. On the suggested view, they would all be comprehended within par. (xxi.). But, in my opinion, but for the presence of par. (xxii.), there would not have been present in the Constitution a power to make laws with respect to divorce and matrimonial causes. In any case, having regard to the presence and terms of par. (xxii.), whatever might otherwise have been comprehended in the legislative power with respect to marriage, in my opinion, a matrimonial cause within par. (xxii.) will not be so comprehended. Matrimonial causes as a whole fall expressly and exclusively within the topic of par. (xxii.), for that paragraph has in reality two topics, "divorce" and "matrimonial causes". Further, only ancillary proceedings in relation to maintenance etc. can constitute matrimonial causes within that paragraph. Therefore, unless proceedings for maintenance, custody or the settlement of property are ancillary to a proceeding for divorce or for nullity of marriage, the Parliament, in my opinion, has no power to create a jurisdiction to entertain them. Thus, it is my opinion that s. 39, read with pars (c), (d), (e) and (f) of the definition matrimonial cause in s. 4(1) and relating to proceedings which are not ancillary to proceedings under pars (a) and (b) of the definition, is not a law with respect to marriage. I would be of the same opinion and for like reasons with respect to ss. 64, 74 and 78. There can be little doubt that the intention of the Parliament was to create a jurisdiction to entertain proceedings of the kind described in pars (c), (d), (e) and (f) of the definition of matrimonial causes, irrespective of whether or not they are ancillary to proceedings within pars (a) and (b) of that definition. For example, on the true construction of ss. 8, 9(4) and 4(2) of the Act, it is evident, in my opinion, that the jurisdiction sought to be created is a jurisdiction to make orders with respect to the property rights of persons who may have been married but presently are not. I have already referred to the attempt to create a jurisdiction with respect to property in which neither party has any right by virtue of marriage but only, for example, by virtue of contract or inheritance. Such proceedings and other proceedings unrelated to proceedings for divorce or nullity are contemplated by the definition in pars (c), (d), (e) and (f) of s. 4(1). The Act in these definitions has purported expressly to create a jurisdiction which embraces the traditional deserted wives and children legislation of the States. That legislation of the States attempts to adjust the rights of people who are still married to each other and to enforce their obligations in respect of their children. Further, courts of the States exercising equitable jurisdiction determine claims to property by citizens who include those who are or have been married. The laws creating and regulating these jurisdictions, in my opinion, could not properly be regarded as laws with respect to marriage. Sections 8(1) and 9(4) evidently intend that all the existing State law with respect to deserted wives and children and, as well, jurisdiction of the equity courts of the States with respect to property claims between married, or erstwhile married, persons shall be wholly superseded. It remains to consider whether s. 39 is invalid in relation to proceedings within pars (c), (d), (e) and (f) of the definition when they are in fact ancillary to proceedings for divorce or nullity. It was submitted by the Solicitor-General for the Commonwealth that the operation of those paragraphs could be confined by construction to proceedings which are so ancillary. But, in my opinion, it is not possible so to construe the relevant sections. The legislative intention to the contrary of such a construction is all too evident. Many provisions of the Act indicate that it was the Parliamentary intention to create a jurisdiction under the Act to make orders for maintenance, for custody, and in respect of property, irrespective of there having been or being current any proceedings for dissolution or annulment of marriage. Sections 8, 9(4), 75, 78 and 81 and other sections of Pt VIII of the Act are so evidently based on that intention that it seems to me that by no process of construction can these paragraphs of the definition be confined to proceedings which are ancillary to proceedings for dissolution or nullity of marriage. However, though it is not possible, in my opinion, by construction to confine s. 39, read with pars (c), (d), (e) and (f) of the definition of "matrimonial cause", to the creation of a jurisdiction to hear and determine proceedings which are ancillary to proceedings under pars (a) and (b) of the definition, I have come to the conclusion that it is possible by the use of s. 15A of the Acts Interpretation Act 1901-1973 to apply the words of s. 39 in combination with pars (c), (d), (e) and (f) only to ancillary proceedings which are within the competence of the Parliament. The process of distributing the operation of s. 39 read with the paragraphs of the definition, though not really achieved in strictness by construction, is a permissible and indeed an appropriate process required by s. 15A of the Acts Interpretation Act. By the use of s. 15A, therefore, s. 39 in conjunction with pars (c), (d), (e), and (f) of the definition will be effective to give and to invest jurisdiction to hear and determine proceedings described in these paragraphs which are ancillary to proceedings under pars (a) and (b) of the definition. Section 39 in conjunction with these paragraphs will be effective to give jurisdiction to hear and determine such ancillary proceedings and proceedings contemplated in par. (f) which relate to ancillary proceedings falling within pars (c) to (e). The courts to which s. 39 refers have, in my opinion, jurisdiction to hear and determine such ancillary proceedings. For the reasons I have expressed, I would answer the questions posed for us by my brother Stephen as follows: 1. (i) No. (ii) Not in relation to proceedings which are not ancillary to principal relief. (iii) Not in relation to proceedings which are neither proceedings for principal relief nor ancillary to proceedings for principal relief. (iv) Not in relation to proceedings which are not ancillary to principal relief. (v) Not in relation to proceedings which are not ancillary to principal relief. (vi) Not in relation to proceedings which are not ancillary to principal relief. 2. Not necessary to answer. 3. (a) Does not arise. (b) Does not arise. (c) Does not arise. The matters should be remitted, pursuant to the request of counsel for the petitioner and applicant and to s. 45 of the Judiciary Act to the respective courts from which the matters have come to this Court, there to be dealt with according to law. The case of Farrelly v. Farrelly may be so remitted by this Court to the Supreme Court of South Australia and the case of Russell v. Russell may be remitted to Stephen J. for remission by him to the Supreme Court of Victoria. Gibbs J. In these two cases, which have been heard together, a challenge has been made to the validity of certain of the provisions of the Family Law Act 1975 Cth ("the Act"). The challenge raises for decision two quite different constitutional questions. The first of those questions is whether the Parliament, having invested the Supreme Courts of the States with federal jurisdiction to hear and determine certain matrimonial causes (s. 39(5) of the Act), may validly proceed to require proceedings in those Courts to be heard in closed court (s. 97(1)) and to direct that neither the judges hearing proceedings under the Act nor counsel shall robe (s. 97(4)). The provisions of s. 97 are not directed only to the Supreme Court; they apply to all courts exercising jurisdiction under the Act, and such jurisdiction may be exercised by the Family Court which the Act sets up, or by the Supreme Courts of the States, or by the courts of summary jurisdiction of the States which by s. 39(6) are in certain classes of cases invested with federal jurisdiction. There has also been referred to the Full Court the question whether s. 123 of the Act is beyond the power of the Parliament to the extent that it authorizes the making of regulations for or in relation to the practice and procedure to be followed in courts other than the Family Court exercising jurisdiction under the Act. That question depends upon considerations similar to those which govern the validity of s. 97. However, it is a question which it is unnecessary to answer for the purposes of the present proceedings. The only regulation made under s. 123 to which any objection has been suggested is reg. 5(2) of the Family Law Regulations, which provides that in proceedings to which s. 9 of the Act applies, the court or a registrar may give directions in all matters of practice and procedure. Section 9, to which later reference will be made, deals with proceedings pending when the Act was passed. It has not been suggested that any direction relevant to the proceedings in either of the cases before us has been given under reg. 5(2). In other words, nothing done under s. 123 has affected the position of any of the parties in either of the cases before us. The decision of the question as to the validity of s. 123 would have no practical consequences affecting the parties to those cases. It will be necessary in the course of discussing s. 97 to say something as to the power of the Parliament to regulate the practice and procedure of courts exercising invested jurisdiction, but it is not necessary to consider the limits of that power or to give a formal answer as to the validity of s. 123. Section 77(iii.) of the Constitution provides that with respect to any of the matters mentioned in ss. 75 and 76 of the Constitution the Parliament may make laws "investing any court of a State with federal jurisdiction". The words of this provision, understood in their natural sense, mean that the Parliament may take an existing State court and clothe it with federal jurisdiction; they contain nothing to suggest that the Parliament may in any way interfere with a court which it endows with jurisdiction. So far as the Parliament is concerned, the courts of the State are, as the description itself implies, the "judicial organs of another Government": Le Mesurier v. Connor [1] ; Bond v. George A. Bond and Co. Ltd. and Bond's Industries Ltd. [2] . They are created and maintained by the States for the purpose of administering the laws in force in the States—a purpose vital to the existence of the States as free communities. They include the Supreme Courts, which were at federation, as they are now, of the highest standing in the community. At the time of federation it would have seemed quite unthinkable that the Commonwealth should be given power to dictate the manner in which those Courts should perform their important functions, even when exercising federal jurisdiction. No power to legislate with respect to those Courts is conferred by the Constitution. 1. (1929) 42 C.L.R. 481, at p. 495. 2. (1930) 44 C.L.R. 11, at p. 22. The "sole source of power to confer Federal jurisdiction on State courts" is s. 77(iii.) of the Constitution: Collins v. Charles Marshall Pty. Ltd. [3] , citing R. v. Federal Court of Bankruptcy; Ex parte Lowenstein [4] . The general powers of the Parliament to legislate with respect to the subject matters confided to it do not authorize legislation giving jurisdiction to State courts: Le Mesurier v. Connor [5] . Of course the power conferred by s. 77(iii.) carries with it whatever is necessary to make that power effective. However, the power to be made effective, as was said by Knox C.J., Rich and Dixon JJ. in Le Mesurier v. Connor [5] , is a power "to confer additional judicial authority upon a Court fully established by or under another legislature". Their Honours continued: Such a power is exercised and its purpose is achieved when the Parliament has chosen an existing Court and has bestowed upon it part of the judicial power belonging to the Commonwealth. To affect or alter the constitution of the Court itself or of the organization through which its jurisdiction and powers are exercised is to go outside the limits of the power conferred and to seek to achieve a further object, namely, the regulation or establishment of the instrument or organ of Government in which judicial power is invested, an object for which the Constitution provides another means, the creation of Federal Courts. Similarly the provisions of s. 51(xxxix.) of the Constitution do not give the Parliament power to legislate with respect to a court which has been invested with federal jurisdiction: see Le Mesurier v. Connor [6] . It is unnecessary for present purposes to consider the questions raised but left open in the judgment in that case [7] —particularly whether a State court invested with federal jurisdiction is part of the "Federal Judicature" within s. 51(xxxix.), a matter upon which conflicting opinions have been expressed. On any view s. 51(xxxix.) does not empower the Parliament to re-fashion an existing judicial organ which it selects as the recipient of a bestowal of federal jurisdiction. 1. (1955) 92 C.L.R. 529, at p. 556. 2. (1938) 59 C.L.R. 556, at p. 586. 3. (1929) 42 C.L.R., at p. 496. 4. (1929) 42 C.L.R., at p. 496. 5. (1929) 42 C.L.R., at pp. 497-498. 6. (1929) 42 C.L.R., at p. 498. In expressing the notion that the Parliament is entitled to do no more than confer additional jurisdiction on an existing court and to make that investiture effective, members of this Court have used a number of expressions to indicate what is beyond the power of the Parliament. Thus it has been said that "the reconstitution of the tribunal itself or of the organization through which its powers and jurisdiction are exercised" is beyond the power of the Parliament: Le Mesurier v. Connor [7] (and see [8] ); the Parliament cannot change "the character or constitution" of the State court or its "nature as a judicial organism": Adams v. Chas. S. Watson Pty. Ltd. [9] ; "the constitution or structure of the court cannot be changed by the Federal Parliament": Peacock v. Newtown Marrickville and General Co-operative Building Society No. 4 Ltd. [10] . It has been argued that the Parliament, in enacting s. 97 of the Act, did not go beyond the limits of its power as indicated by those descriptive phrases. However, expressions such as these were not in my opinion intended to be, and are not, a complete statement of the respects in which legislative power is lacking. They are only attempts to indicate aspects of the broad principle that the court to be invested with jurisdiction is the State court "as it exists": see Peacock v. Newtown Marrickville and General Co-operative Building Society No. 4 Ltd. [10] , and see also Kotsis v. Kotsis [11] . The statement that the Parliament must take the State court as it finds it is obviously too wide if intended to refer to the limits of the court's jurisdiction as defined by the State law, because the power to invest a court with federal jurisdiction necessarily enables the Parliament to enlarge or alter its jurisdiction. Such a statement is, however, in my opinion correct if it is understood to mean that the court itself must be taken as it is found. In Adams v. Chas. S. Watson Pty. Ltd. [12] Latham C.J. spoke of "the distinction between structure and function"; he went on to say: "The Commonwealth Parliament cannot change the structure of a State court, but it may confer new functions upon such a court". Clearly that statement is correct—the exercise of invested federal jurisdiction may entail the performance of new functions. However, it does not follow, and in my opinion Latham C.J. did not mean, that any interference with the functions of a court invested with federal jurisdiction is necessarily within the power of the Parliament. Such a statement would be far too wide and would ignore the principles by which the validity of the legislation is to be decided. The question in my opinion is whether the Parliament has done more than to invest the court with federal jurisdiction and to make such investiture effective. 1. (1929) 42 C.L.R., at p. 498. 2. (1929) 42 C.L.R., at p. 496. 3. (1938) 60 C.L.R. 545, at p. 554. 4. (1943) 67 C.L.R. 25, at p. 37. 5. (1943) 67 C.L.R. 25, at p. 37. 6. (1970) 122 C.L.R. 69, at p. 84. 7. (1938) 60 C.L.R., at p. 555. In Lorenzo v. Carey [13] , five members of this Court cited with approval the remark of Isaacs J. in Baxter v. Commissioners of Taxation (N.S.W.) [14] : "Once the jurisdiction became federal the Commonwealth Parliament could at will regulate the procedure and control the method and extent of relief ". As a generalization this dictum appears to be correct, although the source of the Parliament's authority to regulate procedure in State courts exercising federal jurisdiction is by no means clear. In The Commonwealth v. Limerick Steamship Co. Ltd. [15] , Isaacs and Rich JJ. said that "the procedure and the practice to be observed in relation to Federal jurisdiction is incidental to the powers of the Federal Judicature and so within par. xxxix of sec. 51 of the Constitution "; as I have indicated, it is unnecessary for present purposes to consider whether a State court exercising federal jurisdiction is part of the federal judicature. In other cases it has been suggested that the powers to legislate with respect to the topics mentioned in other paragraphs of s. 51 would provide the necessary authority. In Bond v. George A. Bond and Co. Ltd. and Bond's Industries Ltd. [16] , Rich and Dixon JJ. said that "Section 77(iii.) considered with sec. 51(xvii.) confers ample power upon the Parliament to bestow upon State Courts" (scil., invested with federal jurisdiction in bankruptcy) "all powers appropriate to bankruptcy jurisdiction and all authority incidental to the exercise of such powers". In Kotsis v. Kotsis [17] , Menzies J. apparently took a similar view; he said that s. 127 of the Matrimonial Causes Act 1959 Cth (as amended) (which gave the Governor-General power to make rules for or in relation to the practice and procedure of the courts having jurisdiction under that Act) was a valid exercise of the power of Parliament conferred by s. 51 (xxii.). The questions of the extent of and limits on the power of the Parliament to regulate the practice and procedure of a State court invested with federal jurisdiction were not fully discussed in those cases, and it is not necessary in the present case to express any concluded opinion upon them. It seems clear enough that the Parliament could validly legislate with respect to such matters as the time within which proceedings should be brought (Adams v. Chas. S. Watson Pty. Ltd. [18] ) and the rules of evidence to be applied (Milicevic v. Campbell [19] ). I accept that it might also prescribe rules of practice and procedure to be observed. But on that assumption it is necessary to consider whether a particular provision is no more than a law regulating the practice and procedure which the State court is to follow in exercising its invested jurisdiction, or whether it goes further and is a law with respect to the court itself. 1. (1921) 29 C.L.R. 243, at p. 253. 2. (1907) 4 C.L.R. 1087, at p. 1145. 3. (1924) 35 C.L.R. 69, at p. 105. 4. (1930) 44 C.L.R., at p. 22. 5. (1970) 122 C.L.R., at p. 89. 6. (1938) 60 C.L.R. 545. 7. (1975) 132 C.L.R. 307, at p. 316. In my opinion, s. 97 of the Act is a law of the latter character. The judges of the Supreme Courts, when sitting in court, traditionally wear robes. Some persons consider that robes are outmoded and should be dispensed with; others regard them as a valuable symbol of the antiquity of the honourable traditions of the law and as a means of adding to the dignity of proceedings that are of their nature serious and important. We are not required to join in this controversy. It would not matter if we thought that judges could perform the duties of their office equally well if they were unrobed or that matrimonial proceedings would more appropriately be conducted in an informal atmosphere by judges wearing informal dress. The question to be decided is to be determined by a consideration of the Constitution, and not by our opinions on matters of policy of that kind. If the Parliament considers that matrimonial jurisdiction should be exercised by judges who do not robe, it can create a federal court in which the wearing of robes is proscribed—as it in fact has done. If it chooses to invest a State court with jurisdiction, it has no power to direct the judges of that court to depart from their existing standards of judicial dress. The question may be thought to be of no great importance, but its implications are by no means insignificant. The robes worn by the judges of the Supreme Court are a mark of their status. If the Parliament could legislate as to what the judges shall or shall not wear, it could force upon a court a mode of dress incompatible with its former dignity and status. Such a law might be regarded as changing the nature or character of the court, but whether or not that view would be correct, the law would in my opinion be one with respect to the court itself. In one sense, such a law might be said to be a law as to practice and procedure. The same might be said of a law which prescribed the places where and the hours during which the court must sit, or how the courtroom should be furnished, or what officials should attend the judge in court, or which required the judge to give his judgment within a particular time. Laws of that kind would go beyond merely providing for the practice and procedure to be observed in State courts exercising federal jurisdiction, and would amount to a regulation of the court itself, which is beyond the powers of the Parliament. It is the ordinary rule of the Supreme Court, as of the other courts of the nation, that their proceedings shall be conducted "publicly and in open view" (Scott v. Scott [20] ). This rule has the virtue that the proceedings of every court are fully exposed to public and professional scrutiny and criticism, without which abuses may flourish undetected. Further, the public administration of justice tends to maintain confidence in the integrity and independence of the courts. The fact that courts of law are held openly and not in secret is an essential aspect of their character. It distinguishes their activities from those of administrative officials, for "publicity is the authentic hall-mark of judicial as distinct from administrative procedure" (McPherson v. McPherson [21] ). To require a court invariably to sit in closed court is to alter the nature of the court. Of course there are established exceptions to the general rule that judicial proceedings shall be conducted in public; and the category of such exceptions is not closed to the Parliament. The need to maintain secrecy or confidentiality, or the interests of privacy or delicacy, may in some cases be thought to render it desirable for a matter, or part of it, to be held in closed court. If the Act had empowered the Supreme Courts when exercising matrimonial jurisdiction to sit in closed court in appropriate cases I should not have thought that the provision went beyond the power of the Parliament. In requiring them to sit in closed court in all cases—even proceedings for contempt—the Parliament has attempted to obliterate one of their most important attributes. This it cannot do. 1. [1913] A.C. 417, at p. 441. 2. [1936] A.C. 177, at p. 200. In my opinion the provisions of s. 97 are an attempt to regulate the State courts which are invested with federal jurisdiction. In so far as the provisions of that section relate to courts other than the Family Court they go beyond the powers of the Parliament and are invalid. On this view it is unnecessary to answer the question whether non-compliance with s. 97 by a court other than the Family Court would affect the validity of any judgment made by such court in the exercise of the jurisdiction with which it is invested under the Act. I should, however, perhaps say that my present opinion is that if it be assumed that s. 97 is valid a failure to comply with its provisions would not render a judgment void. Equally, if a judge sat unrobed and in closed court when he should not have done so, the proceedings would not thereby be rendered void. The second broad question that falls for consideration is whether it was within the power of the Parliament to enact in their present form those provisions of the Act that deal with the custody of children, the making of orders for maintenance and the alteration of interests in property. It is necessary to refer briefly to the circumstances of the two cases now before us to indicate exactly what needs to be decided. In Farrelly v. Farrelly the question is whether proceedings brought under the Guardianship of Infants Act, 1940 S.A. by a married woman against her husband for the custody of a child of the marriage and pending when the Act came into operation must be continued and dealt with as if they were proceedings instituted under the Act. In Russell v. Russell the wife petitioned for divorce under the Matrimonial Causes Act 1959 Cth and by her position sought orders for custody of the children, maintenance for herself and for the children and a settlement upon her of her husband's interest in the former matrimonial home. The husband crosspetitioned and disputed her claims for ancillary relief. Subsequently he filed a notice pursuant to s. 9(2) of the Act requesting that further proceedings in the suit be dealt with as if such proceedings had been instituted under the Act on the ground specified in s. 48 of the Act. In these circumstances there have been reserved for our consideration the questions whether ss. 9(4), 39, 64, 74 and 78 of the Act are validly enacted. Before turning to those sections it is necessary to set out the definition of matrimonial cause contained in s. 4(1) of the Act, since that definition and s. 39 together form the keystone of the jurisdictional structure which the Act erects. The definition is as follows: "matrimonial cause" means— (a) proceedings between the parties to a marriage for a decree of— (i) dissolution of marriage; or (ii) nullity of marriage; (b) proceedings for a declaration as to the validity of a marriage or of the dissolution or annulment of a marriage by decree or otherwise; (c) proceedings with respect to— (i) the maintenance of one of the parties to a marriage; (ii) the property of the parties to a marriage or of either of them; or (iii) the custody, guardianship or maintenance of, or access to, a child of a marriage; (d) proceedings between the parties to a marriage for the approval by a court of a maintenance agreement or for the revocation of such an approval or for the registration of a maintenance agreement; (e) proceedings for an order or injunction in circumstances arising out of a marital relationship; or (f) any other proceedings (including proceedings with respect to the enforcement of a decree or the service of process) in relation to concurrent, pending or completed proceedings of a kind referred to in any of paragraphs (a) to (e), including proceedings of such a kind pending at, or completed before, the commencement of this Act. The term "maintenance agreement", which is used in par. (d) of the definition of "matrimonial cause", is widely defined in s. 4(1) and includes an agreement in writing made between the parties to a marriage that makes provision with respect to financial matters. Proceedings of the kind referred to in par. (a) or (b) of the definition of "matrimonial cause" are "proceedings for principal relief": s. 4(1). By s. 8(1), after the commencement of the Act (a) proceedings by way of a matrimonial cause shall not be instituted except under the Act, and (b) proceedings by way of a matrimonial cause instituted before the commencement of the Act shall not be continued except in accordance with s. 9. By s. 9(1) pending proceedings for (inter alia) a decree of dissolution of marriage may be continued and shall be dealt with as if the Act had not been passed. That provision is subject to s. 9(2), which enables a party, in the circumstances mentioned in that sub-section, to request that pending proceedings for dissolution of marriage shall be dealt with as if they were proceedings instituted under the Act on the ground referred to in s. 48. By s. 9(4) it is provided as follows: Pending proceedings constituting a matrimonial cause, not being proceedings for principal relief, whether instituted under the repealed Act or under the law of a State or Territory, may be continued and shall be dealt with as if they were proceedings instituted under this Act. Section 39, which has already been mentioned, provides that matrimonial causes may be instituted in the Family Court or in the Supreme Court of a State or Territory or, in the case of proceedings which are not proceedings for principal relief, in a court of summary jurisdiction of a State or Territory. Section 64 deals with the powers and duties of a court in custody proceedings. Section 74 provides that in proceedings with respect to the maintenance of a party to a marriage or of a child of a marriage the court may make such order as it thinks proper for the provision of maintenance in accordance with Pt VIII of the Act. Section 78(1) provides that in proceedings between the parties to a marriage with respect to existing title or rights in respect of property the court may declare the title or rights, if any, that a party has in respect of the property. By s. 78(2) the court may make consequential orders, including orders as to sale, partition and possession. By s. 78(3) an order under the section is binding on the parties to the marriage but not on any other person. In the course of argument we were referred to a number of other sections which show the intended width of the scope of the Act. Section 4(2) provides that a reference to a party to a marriage includes a reference to a person who was a party to a marriage that has been dissolved or annulled. By s. 5(1) certain children are deemed to be children of a marriage. Section 72 deals with the duty of one party to a marriage to maintain the other and the provisions of this section would have an extended operation in the light of s. 4(2). Section 79 provides that in proceedings with respect to the property of the parties to a marriage or either of them the court may make such order as it thinks fit altering the interests of the parties in the property. Section 79(3) provides that the court shall not make an order under this section unless a decree for dissolution or nullity has been made or proceedings for a decree of dissolution or nullity have been instituted or a party has filed in the court a notice under s. 15. Under s. 15 a party may file a notice at will. However, I find it unnecessary to deal with the validity of these sections for the purposes of the matters now before the Court. That can be considered when it becomes necessary to do so for the purpose of deciding the rights of parties affected by the provisions of those sections. It is clear from the definition of "matrimonial cause" that proceedings of the kind referred to in par. (c) constitute a "matrimonial cause" even if they are not ancillary to proceedings for principal relief. It is clearly the intention of the Act, for example, that proceedings brought between parties to a marriage for an order respecting the custody of a child must be instituted or continued, as the case may be, under the Act although they were not associated with any claim for dissolution of marriage or other proceedings for principal relief. It is necessary to consider whether provisions having this effect were within the legislative power of the Parliament. By s. 51 of the Constitution the Parliament is empowered to make laws with respect to— (xxi) Marriage: (xxii) Divorce and matrimonial causes; and in relation thereto, parental rights, and the custody and guardianship of infants: It has been settled by the decision in Attorney-General (Vict.) v. The Commonwealth [22] that the power given by par. (xxi.) is not limited to making laws with respect to the solemnization of marriage. The word "marriage" in the paragraph is not to be understood as having the narrow meaning of act or ceremony of marriage; it means the relationship that exists between man and wife. The power to make laws with respect to that relationship does of course enable the Parliament to legislate as to the manner in which the relationship may be created, but it extends also to the making of laws which provide for the legal consequences that flow from the relationship, at least laws as to the rights and obligations of the parties to the marriage that arise out of the marriage relationship. This view was, I think, accepted by the majority of the Court in Attorney-General (Vict.) v. The Commonwealth —see per Kitto J. [23] ; per Taylor J. [24] ; per Menzies J. [25] ; and per Owen J. [26] . Kitto J. left open the question whether a law operating by reference to the marriage status, such as a Married Women's Property Act, would be a law with respect to marriage [23] and it is unnecessary to decide that question in the present case. Unless par. (xxii.) limits the meaning of par. (xxi.)—a question to which I shall shortly turn— the power to make laws with respect to marriage would in my opinion enable the Parliament to declare the duty of one party to a marriage to support and maintain the other. Since the procreation and nurture of children is one of the ends of marriage, the power would also in my opinion enable Parliament to declare the rights and duties of the parties to a marriage with respect to the custody and maintenance of the children of the marriage. When the Parliament is empowered by one of the paragraphs of s. 51 to create substantive rights and liabilities it may also under the same power provide for the enforcement of those rights in legal proceedings, although the investiture of a State court with federal jurisdiction for that purpose is to be effected under s. 77(iii.). Therefore par. (xxi.), if read alone, would in my opinion enable the Parliament to legislate for the enforcement of the rights which one party had against the other and which arose from the marriage relationship, including rights to maintenance and custody. 1. (1962) 107 C.L.R. 529. 2. (1962) 107 C.L.R., at p. 554. 3. (1962) 107 C.L.R., at p. 560. 4. (1962) 107 C.L.R., at p. 572. 5. (1962) 107 C.L.R., at p. 602. 6. (1962) 107 C.L.R., at p. 554. However, par. (xxii.) of s. 51 expressly gives the Parliament power to legislate with respect to divorce and matrimonial causes. The latter expression, if widely understood, would refer to any controversy between the parties to a marriage as to a matter which pertained to the marriage relationship. It would indeed include divorce itself. If it were not for the concluding words of par. (xxii.), that paragraph could without difficulty be read as supplementing and amplifying, so far as necessary, the power given by par. (xxi.); together the two paragraphs would give the fullest power to legislate with respect to proceedings brought by one spouse against another to enforce any rights which had their source in the matrimonial relationship. However, the concluding words of par. (xxii.), by giving a power to make laws with respect to parental rights and the custody and guardianship of infants in relation to divorce and matrimonial causes, indicate a clear intention that the power given by par. (xxii.) should not authorize legislation with respect to those questions unless they arise as an incident to proceedings for divorce or some other matrimonial cause. Under par. (xxii.) the Parliament has power to deal with proceedings for custody which are brought as ancillary to proceedings for divorce or some other matrimonial cause, but has no power to deal with custody proceedings brought independently of any other claim to relief. The question that then arises is whether this limitation affects not only the power conferred by par. (xxii.) but also that granted by par. (xxi.). It has been said not infrequently that as a general rule the paragraphs of s. 51 should not be read as limiting each other in any way. But that is only a general rule; it admits of exceptions. One exception to it that is already clearly established is that the powers given by paragraphs of s. 51 other than par. (xxxi.) cannot be interpreted as enabling the Parliament to legislate with respect to the acquisition of property except on just terms: Re Dohnert Müller Schmidt and Co.; Attorney-General (Cth) v. Schmidt [27] . Paragraphs (xxi.) and (xxii.) are placed together in s. 51 and they deal with subjects that are very closely associated and that to some extent quite obviously overlap. This strongly suggests that one cannot be disregarded in construing the other. In Attorney-General (Vict.) v. The Commonwealth [28] the argument that par. (xxi.) should be read as referring only to the solemnization of marriage was supported by the suggestion that par. (xxii.) authorized laws with respect to the dissolution of marriages and that this was a power which would have been conferred by par. (xxi.) if it had stood alone. As to this argument, Taylor J. said [29] : I do not assent to this contention and would not be prepared to do so even if it seemed to me that, standing alone, par. (xxi.) would justify the enactment of every law expressly authorized by par. (xii.) for the fact that the constitutional instrument contains express provision for the matters mentioned in the latter paragraph provides no warrant for reading "Marriage", merely, as "Solemnization of Marriages". What must be borne in mind is that the expression with which we are concerned is used to define a broad constitutional power and in the paragraph in question the word "marriage"—appearing without limitation or qualification—is entitled to as wide an interpretation as it can reasonably bear. Menzies J. said [30] : Although I am disposed to think that had there been no s. 51(xxii.) the marriage power would of itself have covered divorce, I do not think the existence of s. 51(xxii.) requires the implication that s. 51(xxi.) is limited to determining who may marry and the forms and ceremonies of marriage. These observations do not in my opinion assist in considering the question whether the concluding provision of par. (xxii.), containing as it does words intended to indicate a limitation on the power of Parliament, operates to limit the power conferred by par. (xxi.) as well as that conferred by par. (xxii.). In Lansell v. Lansell [31] , where the Court upheld the provisions of s. 86 of the Matrimonial Causes Act 1959 Cth (which enabled the court to require one party to a marriage to make a settlement of property in favour of the other), it was emphasized in the judgments of all the members of the Court that the jurisdiction created by the section could only be exercised in proceedings for divorce or some other matrimonial cause, i.e. where the order made was ancillary or incidental to such proceedings [32] . Kitto and Menzies JJ. expressly stated their doubts whether if the section had created a jurisdiction that could be exercised independently of divorce or other matrimonial proceedings it would have been within power [33] . In my opinion it is not proper in the construction of par. (xxi.) to ignore the restrictions on power contained in par. (xxii.). To do so would in effect make the concluding words of par. (xxii.) quite ineffective. Paragraph (xxii.) resembles par. (xxxi.) in that in each case there is an express indication of an intention that the power of the Parliament should be subject to a specified limitation. It would in my opinion give altogether too little weight to the words of par. (xxii.) to regard par. (xxi.) as granting a power to make laws with respect to parental rights and the guardianship and custody of infants even when those matters had no relation to divorce or any other matrimonial cause. 1. (1961) 105 C.L.R. 361, at pp. 370-372. 2. (1962) 107 C.L.R. 529. 3. (1962) 107 C.L.R., at p. 560. 4. (1962) 107 C.L.R., at p. 572. 5. (1964) 110 C.L.R. 353. 6. (1964) 110 C.L.R., at pp. 359, 367, 368, 369. 7. (1964) 110 C.L.R., at pp. 359, 368. I have already indicated that the combined effect of ss. 8(1)(b), 9(4) and 39 of the Act, read in the light of par. (c)(iii) of the definition of matrimonial cause in s. 4, is that proceedings for the custody of a child, brought by one spouse against another under State law but independently of any divorce or other matrimonial cause, are to be continued under the Act. For the reasons I have given, the powers conferred by pars (xxi.) and (xxii.) of s. 51 of the Constitution do not empower the Parliament to enact provisions having that effect. In so far as the Act requires the proceedings in Farrelly v. Farrelly to be continued under its provisions, it is beyond the power of the Parliament. The proceedings in Russell v. Russell are for divorce and ancillary relief. It was clearly competent for the Parliament to legislate with respect to such proceedings. The question that arises is whether because some of the provisions of the Act have a wider application than is constitutionally permissible the whole of the Act, or at least the vital s. 39, is in consequence invalid. There are other respects than those which I have already discussed in which the Act may go beyond power. For example, the combined effect of par. (c)(ii) of the definition of "matrimonial cause", and of ss. 4(2), 39 and 78 appears to be that the Act gives jurisdiction to entertain proceedings brought between persons no longer married and relating to rights of property which did not arise out of the matrimonial relationship. Paragraph (d) of the definition of "matrimonial causes", considered in the light of the definition of "maintenance agreement", might include proceedings as to matters unconnected with the matrimonial relationship, and pars (e) and (f) of that definition include proceedings which are not proceedings between the parties to the marriage. I have already referred to the provisions of s. 5 which in conjunction with s. 39 and s. 64 purport to apply the provisions of the latter sections to proceedings with respect to the custody of a child who is in truth not the child of a marriage. However, it seems to me to be unnecessary for the decision of the present matters to determine how far the Act goes beyond power. All that is necessary to decide is whether the provisions of the Act have a valid operation in so far as they relate to proceedings for divorce and ancillary relief between the parties to a marriage. Section 39 of the Act confers a jurisdiction in matrimonial causes which by embracing all the proceedings described in the definition of that expression is too wide. It is the second class of case referred to by Dixon J. in his judgment in R. v. Poole; Ex parte Henry [No. 2] [34] . The question therefore is whether the Parliament intended the provisions to have a distributive effect. The provisions of s. 15A of the Acts Interpretation Act 1901-1973 Cth create the presumption that it did have that intention, that is, that the general words were intended to be distributable: Bank of New South Wales v. The Commonwealth [35] . If the definition in par. (c) is confined to proceedings between the parties to the marriage which are ancillary to proceedings for principal relief, s. 39, operating in conjunction with the provisions of par. (c) of the definition, will be within power. It would not, in my opinion, be sufficient to confine par. (c) to proceedings between the parties to a marriage not being ancillary proceedings, because for the reasons already given the provisions of pars (ii) and (iii) of par. (c) of the definition would then both go beyond power. The only way in which all the general provisions of par. (c) can be limited so as to confine them within constitutional bounds is to restrict the definition to proceedings between the parties to a marriage being ancillary to proceedings for principal relief. Section 9(4) of the Act needs to be similarly confined in order to render it valid. The intention of the Parliament as expressed in s. 15A of the Acts Interpretation Act should be given effect by restricting the proceedings referred to in par. (c) of the definition of matrimonial cause in s. 4 of the Act to proceedings between the parties to the marriage which are ancillary to proceedings for principal relief. When the effect of the Act is limited in this way no problem arises as to the operation of ss. 64, 74 and 78 because those sections can only apply in cases in which a court has jurisdiction under s. 39. 1. (1939) 61 C.L.R. 634, at p. 652. 2. (1948) 76 C.L.R. 1, at p. 371. I would make the following orders: Farrelly v. Farrelly. I would remit the case to the Supreme Court of South Australia to be heard and determined in accordance with the provisions of the Guardianship of Infants Act, 1940 S.A.. Russell v. Russell. I would answer the questions referred to the Full Court as follows: 1. (i) No. (ii) Not in relation to proceedings which are not ancillary to proceedings for principal relief. (iii) Not in relation to proceedings which are neither proceedings for principal relief nor ancillary to proceedings for principal relief. (iv), (v), (vi) Not in relation to proceedings which are not ancillary to proceedings for principal relief. 2. Not answered. 3. Not answered. I would remit the proceedings to the Supreme Court of Victoria to be heard and determined in accordance with the Act. Stephen J. Two quite distinct questions of constitutional validity are raised in these two cases. The more far-reaching arises in both, the narrower arises only in Russell v. Russell. As to the former, concerned with the extent of legislative power conferred by pars (xxi.) and (xxii.) of s. 51 of the Constitution, I have read and agree with the reasons for judgment of Mason J. I would accordingly answer pars (ii) to (vi) of question 1 and question 2 as he does. I turn, then, to what I have described as the narrower question, whether the Commonwealth Parliament, in the course of investing State Supreme Courts with federal jurisdiction in matrimonial causes, may validly require that all proceedings in those Courts should take place "in closed court"—Family Law Act s. 97(1), and that "Neither the judge nor counsel shall robe" —s. 97(4). The source of express power to invest with federal jurisdiction lies in s. 77(iii.) of the Constitution and the answer to this question turns upon whether or not the grant of power in s. 77(iii.) to "make laws investing any court of a State with federal jurisdiction" carries with it, as incidental to the power to invest with jurisdiction, the power to require invested jurisdiction to be exercised without robes and in closed court. The extent of incidental power, whether arising by implication from the grant of express power or from the terms of s. 51 (xxxix.)—as to which compare Le Mesurier v. Connor [36] , and Queen Victoria Memorial Hospital v. Thornton [37] —will be affected by the nature of the subject matter of the express grant which is in question. The nature of the grant of power to invest with federal jurisdiction is now well established. It is not a power to legislate generally "with respect to" the subject matter of such investiture (Le Mesurier v. Connor [38] ), nor does it give power to affect "the organization of State courts" [39] ; while the function of a State court may validly be affected by laws enacted under s. 77(iii.) the court's organization and constitution are inviolate, a distinction being drawn for this purpose between structure and function (Adams v. Chas. S. Watson Pty. Ltd., per Latham C.J. [40] ). In Peacock v. Newtown Marrickville and General Co-operative Building Society No. 4 Ltd. [41] , the Chief Justice put it thus: This is a power to give new, additional, jurisdiction to State courts. The Parliament may select such State courts as it pleases. It may give them much or little new jurisdiction. It may make the jurisdiction as wide or as narrow as it pleases with respect to persons, localities or amounts involved; or, as in the Judiciary Act 1903-1940, s. 39, it may allow the State law to operate in respect of such matters. But the State court must be taken as it exists. The constitution or structure of the court cannot be changed by the Federal Parliament. 1. (1929) 42 C.L.R. 481, at p. 498. 2. (1953) 87 C.L.R. 144, at p. 151. 3. (1929) 42 C.L.R., at p. 500. 4. (1929) 42 C.L.R., at p. 496. 5. (1938) 60 C.L.R. 545. 6. (1943) 67 C.L.R. 25, at p. 37. To establish the validity of sub-ss. (1) and (4) of s. 97 it will not be enough merely to conclude that their subject matter does not intrude upon this prohibited area, the constitution and organization of State courts; it must also appear that they are so germane to the jurisdiction which is being invested as to be incidental to that investiture. It will be convenient to deal with each of the sub-sections in turn, asking as to each whether it satisfies these two requirements. I take first sub-s. (4), the requirement as to robes. The requirement that robes should not be worn does not, in my view, touch upon the constitution or organization of a court. In McPherson v. McPherson [42] , their Lordships observed that the effect of the absence of robes was to make proceedings "in one respect less formal than those of an ordinary trial"; but robed or unrobed the tribunal in question remains unaltered in its constitution and general organization. The widespread, if not universal, absence in Australian jurisdictions of any specification, statutory or otherwise, of what robes shall be worn is indicative of how far removed is this topic from matters affecting the constitution of Supreme Courts; when, as has occurred from time to time in certain Australian jurisdictions, a radical change has occurred in the robes worn on the bench this no more effected a change in the court's constitution than did the donning of wigs by the South Australian Supreme Court bench for the first time in the 1850s or their adoption of the judicial scarlet in the 1870s: Hannan's Life of Chief Justice Way (1960), p. 101. What a bench of judges customarily wears or requires counsel to wear at any particular time forms no part of the constitution or organization of that court. 1. [1936] A.C. 177, at p. 198. On the other hand such a law is sufficiently connected with the particular investing of jurisdiction here in question as to render it incidental to that investiture. The matrimonial causes jurisdiction not only involves the most intimate personal relationships of the parties but is also concerned with status, with the institution of marriage and with parental responsibility, three matters of considerable public concern. It was these latter considerations which led their Lordships in McPherson's Case [43] so strongly to deprecate the informality which they found to surround undefended divorce proceedings in Alberta; however it is no doubt out of regard for the former consideration, the essentially intimate and personal nature of such proceedings, that the legislature, in s. 97, has now stipulated for that very informality which their Lordships found so undesirable. 1. [1936] A.C. 177. Despite the stark contrast between these two approaches to the matrimonial causes jurisdiction, due no doubt in large measure to the changes in community attitudes over the intervening forty years, they are in complete accord in their recognition of what is the effect which the wearing of wig and gown has upon proceedings. If one aim of the Family Law Act is, so far as possible, to dispense with formality it is, I think, properly incidental to the investing of jurisdiction in State Supreme Courts to require that their proceedings shall themselves be free of the formality which wig and gown confer. Accordingly I conclude that s. 97(4) is validly authorized by s. 77(iii.). Different considerations affect s. 97(1), if only because it is concerned not with mere curial dress but with a matter of great substance, the concept of the hearing in open court. It would be an unnecessary and profitless digression to attempt any account of the long history and high significance attaching to open hearings in English courts of justice; it is all most eloquently exposed in the judgment of the members of the Full Court of Appeal in Scott v. Scott [44] and in the speeches of their Lordships in the appeal to the House of Lords [45] . 1. [1912] P. 241. 2. [1913] A.C. 417. What I infer from all that was said in Scott v. Scott is that a tribunal which as of course conducts its hearings in closed court is not of the same character as one which habitually conducts its proceedings in open court. It is one of the "ordinary incidents of English courts of justice" that its proceedings should be conducted in public (per Bramwell B. and per Williams J., H (falsely called C) v. C [46] ), it being "the primary function of the court to administer equal justice to all suitors in open court" (per Farwell L.J., Scott v. Scott [47] ). Viscount Haldane L.C. said in that case, on appeal [48] , that, subject to three well established but only apparent exceptions, courts of justice must, as between parties, administer justice in public; the Earl of Halsbury [49] spoke of every court of justice being open to every subject of the King, and Lord Shaw [50] denied "that it was open to the judges of England to turn their courts into secret tribunals". In Dickason v. Dickason [51] , Barton A.C.J., speaking for the Court, referred to Scott v. Scott [52] and described the admission of the public to attend proceedings as "one of the normal attributes of a court". 1. (1859) 29 L.J. (P. & M.) 29, at p. 30. 2. [1912] P., at p. 287. 3. [1913] A.C., at p. 437. 4. [1913] A.C., at p. 440. 5. [1913] A.C., at p. 481. 6. (1913) 17 C.L.R. 50, at p. 51. 7. [1913] A.C. 417. To require that a Supreme Court, possessing all the attributes of an English court of justice, should sit as of course in closed court is, I think, in the words of Lord Shaw, to turn that Court into a different kind of tribunal and involves that very intrusion into its constitution and organization which s. 77(iii.) does not authorize. It is nothing to the point that there customarily exists power, both as a matter of general law and by statute, for State Supreme Courts to sit in camera; the three types of cases in which, at general law, cases may be heard in camera, those concerned with wardship and lunacy and those involving secret processes, do but constitute exceptions proving the rule; the reason for their existence as apparent exceptions is fully dealt with in the judgments in Scott v. Scott [52] . The fact that in all other instances statutory authority is required for in camera hearing emphasizes that no mere matter of practice or procedure is in question, such as might depend upon the discretion of the judge or upon rules of court; parliamentary sanction is required for the exclusion of the public from proceedings. 1. [1913] A.C. 417. Accordingly I conclude that s. 97(1) of the Family Law Act is beyond power; I may add that but for the view to which I have come concerning this intrusive nature of s. 97(1) I would have found it to be within power since I regard the sub-section as otherwise within the incidental power which is to be associated with the grant of power conferred by s. 77(iii.). In conclusion upon s. 97 I refer to three decisions of this Court. In Baxter v. Commissioner of Taxation (N.S.W.) [53] , Isaacs J. said of s. 39(2) of the Judiciary Act, which invests State Supreme Courts with federal jurisdiction, subject to certain conditions and restrictions: Once the jurisdiction became federal the Commonwealth Parliament could at will regulate the procedure and control the method and extent of relief, and, indeed, under s. 79 of the Constitution could even prescribe the number of Judges by whom the invested jurisdiction should be exercised. In Lorenzo v. Carey [54] five members of the Court expressly adopted these views of Isaacs J. and in The Commonwealth v. Limerick Steamship Co. Ltd. [55] Isaacs and Rich JJ. said of invested jurisdiction that Parliament might invest it as it pleased, "the procedure and the practice to be observed in relation to Federal jurisdiction is incidental to the powers of the Federal Judicature and so within par. xxxix. of s. 51 of the Constitution ", and see per Starke J. [56] . Although what may be worn by judges and by counsel does not, I think, precisely fall within the subject matter either of practice or of procedure of a court, yet I would regard these passages as, at the least, lending some support to the conclusion I have expressed concerning s. 97(4) of the Family Law Act. They cast, I think, no light upon the problem of s. 97(1) if I am correct in treating a general requirement that all proceedings be closed to the public as going not merely to practice or procedure but rather as affecting the very nature of the tribunal in question. 1. (1907) 4 C.L.R. 1087, at p. 1145. 2. (1921) 29 C.L.R. 243. 3. (1924) 35 C.L.R. 69, at p. 105. 4. (1924) 35 C.L.R., at p. 115. In conclusion I pass briefly to consider question 3, which asks whether non-compliance with certain provisions of s. 97 of the Act by State Supreme Courts affects the validity of decrees or orders made in exercise of jurisdiction under the Family Law Act. One of the provisions in question is that contained in sub-s. (3) of s. 97 but as to the consequences of non-compliance with it I say nothing; it requires a court to proceed "without undue formality" and to "endeavour to ensure that the proceedings are not protracted". These requirements seem, at least in the abstract, to lack precise meaning; whether they are capable of being rendered more than merely precatory may emerge if they ever have to be applied to particular facts. In the meantime I refrain from answering question 3 so far as it relates to this sub-section; nor need it be answered in relation to sub-s. (1), which I regard as invalid. Non-compliance with the valid provisions of sub-s. (4), as to judge and counsel being robed, will not, I think, result in invalidity of proceedings. Such a consequence would be as bizarre as it would be unjust and finds no support in authority. The presence of robes in no way alters the substance of the proceedings, it does not go to jurisdiction nor does it effect the legal rights being adjudicated upon the proceedings. To apply to this case the traditional distinction between mandatory provisions and those which are merely directory is inappropriate if the term directory is used to suggest that the legislature required no more than substantial, rather than precise, compliance with its command; in the present case I have no doubt that what is called for is precise compliance. But the contrast between mandatory and directory is in point if it is to the consequences of non-compliance that attention is directed. It cannot, I think, have been contemplated by the legislature that the consequence of non-compliance should be nullity of proceedings, the less so since the non-compliance is not that of a party to the proceedings but of the court itself. This is such a case as was spoken of in the joint judgment of the majority in Clayton v. Heffron [57] ; it is the mode of performance of a public duty, not the exercise of private rights, that is in question and a departure from the required mode of performance should not readily be treated as bringing with it invalidity with all its dire consequences, especially where status and capacity are in question; and see Montreal Street Railway Co. v. Normandin [58] and Ex parte Tasker [59] . I do not think that the observation of their Lordships in Public Prosecutor v. Oie Hee Koi [60] , that enactments regulating the procedure to be followed in courts are usually imperative and not directory, was intended to extend to a case such as the present, so very different from that which their Lordships had in hand; nor does anything in the legislation as a whole conduce to such a mandatory construction. Indeed the vagueness of the immediately proceeding sub-section, to which I have already referred, points strongly in an opposite direction. Accordingly, I conclude that non-compliance with sub-s. (4) produces no effect upon the validity of decrees or orders. 1. (1960) 105 C.L.R. 214, at p. 247. 2. [1917] A.C. 170, at p. 174. 3. [1971] 1 N.S.W.L.R. 804, at p. 808, 815-816. 4. [1968] A.C. 829, at p. 852. I answer questions 1(ii) to (vi) and question 2 in the same way as does Mason J. I answer the remaining questions as follows:— Question 1(i)— as to sub-s. (1) of s. 97, No; as to sub-s. (4) of s. 97, Yes. Question 3(a)—does not arise; 3(b)—no answer given; 3(c)— No. Mason J. It has often been said that the power conferred by s. 77(iii.) of the Constitution on the Commonwealth Parliament to invest a State court with federal jurisdiction is subject to the limitation that Parliament cannot change the constitution, the structure or the organization of the court (Le Mesurier v. Connor [61] ; Kotsis v. Kotsis [62] ). The power is confined to investing jurisdiction, but within its confines, it carries with it "whatever is necessary to give effect to the power itself" (Le Mesurier v. Connor [61] ). 1. (1929) 42 C.L.R. 481, at p. 496. 2. (1970) 122 C.L.R. 69. 3. (1929) 42 C.L.R. 481, at p. 496. The relationship which exists between the general powers of the Parliament to legislate under s. 51 and the particular power conferred by s. 77(iii.) has not been comprehensively considered. However, it was accepted in the last-mentioned case that, by reason of the existence of ss. 77 and 79, "the general powers of the Parliament to legislate must not be interpreted as authorizing legislation giving jurisdiction to State Courts". It has also been accepted that the Commonwealth Parliament may in the exercise of its legislative powers regulate the procedures to be followed by State courts exercising federal jurisdiction (see Baxter v. Commissioners of Taxation (N.S.W.) [63] ; Lorenzo v. Carey [64] ; The Commonwealth v. Limerick Steamship Co. Ltd. [65] ; Adams v. Chas. S. Watson Pty. Ltd. [66] ). For the most part these references do not identify the precise source of the legislative power attributed to the Parliament, though Isaacs and Rich JJ., in The Commonwealth v. Limerick Steamship Co. Ltd. [65] found it in s. 51(xxxix.) of the Constitution. For my part, I should be inclined to think that the general powers to legislate in s. 51 confer authority to enact procedural as well as substantive rules of law upon the topics there enumerated and that a vesting under s. 77(iii.) in a State court of federal jurisdiction in matters arising under a Commonwealth statute gives that court authority and imposes upon it a duty to hear and determine those matters according to the substantive and procedural rules thereby enacted without the need for co-operative assistance from s. 51(xxxix.). However, this is a matter which need not be explored, for the existence of the legislative power to regulate the procedure to be followed in the exercise of federal jurisdiction by State courts has been frequently and authoritatively asserted in the past. 1. (1907) 4 C.L.R. 1087, at p. 1145. 2. (1921) 29 C.L.R. 243, at p. 253. 3. (1924) 35 C.L.R. 69, at p. 105. 4. (1938) 60 C.L.R. 545. 5. (1924) 35 C.L.R. 69, at p. 105. That the provisions contained in s. 97(1) and 97(4) of the Family Law Act requiring the court to sit as a "closed court" and requiring the judge and counsel not to wear robes are procedural rules does not admit of any doubt. It follows that they are valid in their application to State courts exercising federal jurisdiction unless their effect is to alter the constitution, structure or organization of these courts. It should be added that the two provisions are procedural rules which have a special relevance to the jurisdiction in matters of family law vested by the Act in State courts. The absence of robes might well be thought to diminish the air of formality which would otherwise prevail in the proceedings. Likewise, as family law matters are essentially private controversies, there are reasons for thinking that they are more appropriately heard in a closed, rather than an open, court. In each instance the rule of procedure adopted may put the parties and the witnesses more at ease and inspire greater confidence in the judicial process. In my view, neither provision interferes with the constitution, structure or organization of State courts. The word "organization" is not derived from the Constitution. In the joint judgment in Le Mesurier v. Connor [67] it expressed the notion that the staff of the State court, the duties of staff and the relationship between staff and the State court were all subjects beyond the reach of Commonwealth power. Thus it was said that "the reconstitution of the tribunal itself or of the organization through which its powers and jurisdiction are exercised cannot be considered to be a matter incidental to the exercise of the jurisdiction which it takes by the grant" [68] . The word has therefore a limited significance which has sometimes been expressed by the word "structure" used on its own, as by Latham C.J. in Adams' Case [69] : "The Commonwealth Parliament cannot change the structure of a State court, but it may confer new functions upon such a court." 1. (1929) 42 C.L.R. 481. 2. (1929) 42 C.L.R., at p. 498. 3. (1938) 60 C.L.R., at p. 555. Whether the word "organization" in this context should hereafter be taken to include the relationship between a State court and those who appear professionally before it is not a question for present consideration. At this time it is sufficient to say that the wearing of robes does not constitute part of the organization of the court in the narrow sense in which the word has been used. Still less can it be said that sitting in "open court" forms part of a court's organization. No doubt a hearing in open court has been traditionally regarded as of great importance (see Scott v. Scott [70] ). Yet this in itself does not warrant the conclusion that a hearing in open court is a constituent element in the organization of State courts in the sense already explained. 1. [1913] A.C. 417. I would therefore hold that ss. 97(1) and 97(4) are valid. The remaining questions to be answered turn largely on the extent of the legislative powers conferred by ss. 51(xxi.) and (xxii.). It is convenient in the first instance to examine s. 51 (xxii.), curiously expressed as it is, "Divorce and matrimonial causes; and in relation thereto, parental rights, and the custody and guardianship of infants." Although there is much to commend the view that the expression "matrimonial causes", had it stood on its own, would have signified any legal proceedings between the parties to a marriage touching the marriage relationship, the conjunction of the expression with the other words in the paragraph makes it difficult to assign to it that broad meaning. The reference to "custody" might suggest that custody proceedings would not otherwise be included in "matrimonial causes". Moreover in Lansell v. Lansell [71] , Menzies J. regarded the juxtaposition of "Divorce" and "matrimonial causes" as indicating that the expression had a restricted meaning. The decision in Lansell v. Lansell affirming the validity of s. 86 of the Matrimonial Causes Act 1959 proceeded on the narrow ground that the power which it conferred upon the court to make a settlement of the property of the parties to a marriage was exercisable only by way of ancillary relief in proceedings for dissolution of marriage or some other form of principal relief. 1. (1964) 110 C.L.R. 353, at p. 368. In the light of these considerations the jurisdiction conferred by s. 39 of the Family Law Act in a "matrimonial cause" cannot be wholly supported as an exercise of the s. 51(xxii.) power. The statutory definition of "matrimonial cause" contained in pars (c), (d), (e) and (f) of s. 4(1) of the Act plainly provides that proceedings for maintenance, property settlements, guardianship, custody and access, instituted not by way of ancillary relief but independently and separately from proceedings for annulment or dissolution of marriage, constitute "matrimonial causes" with the consequence that jurisdiction in relation to them is conferred by s. 39. This conferment of jurisdiction is not authorized by s. 51(xxii.). Whether the provisions are capable of being read down is a question better left for later consideration. The judgments in Attorney-General (Vict.) v. The Commonwealth [72] (the Marriage Act Case) establish that the power conferred by s. 51(xxi.) should not be narrowly construed [73] . In that case the members of the Court were unanimous in rejecting the notion that the power was confined to the celebration of marriage. Once this step is taken it is obvious that the power extends to the institution of marriage, to the condition or state of being married and, as Taylor, Menzies and Owen JJ. were disposed to hold, to the definition of the respective rights, duties and obligations of the parties arising out or in consequence of marriage [74] . Moreover, subject to the provisions of Ch. III relating to the defining and the investing of jurisdiction in federal and State courts and to any limitation to be derived from the inter-relationship between pars (xxi.) and (xxii.)—a matter shortly to be considered—it will extend to the enforcement of the rights, duties and obligations thereby created without any need to resort to s. 51(xxxix.). The notion that the creation of rights arising out of the marriage relationship stands within s. 51(xxi.) but the enforcement of those rights stands outside the power makes an arbitrary distinction which should be rejected, except in so far as it is necessary to acknowledge that Ch. III makes special provision for defining and investing jurisdiction in courts. 1. (1962) 107 C.L.R. 529. 2. (1962) 107 C.L.R., at pp. 539-540, 560, 577. 3. (1962) 107 C.L.R., at pp. 560, 572, 602. In an endeavour to reduce the content of the marriage power two arguments were advanced. First it was said that it was a power with respect to marriage, not with respect to married persons. This statement is correct so far as it goes, but it fails to acknowledge, as it should, that a law about married persons may also be a law about marriage. A law which defines the rights, duties and obligations of the parties to a marriage and prescribes the consequences in law of the institution of marriage is a law which does not cease to be a law on the topic of marriage merely because it can also be described as a law about married persons. Then it was suggested that the scope of the marriage power was diminished by the existence of s. 51(xxii.) and especially by the phrase "and in relation thereto, parental rights, and the custody and guardianship of infants", a view which was rejected by several members of the Court in the Marriage Act Case (per Taylor J. [75] ; per Menzies J. [76] ). The argument pays insufficient attention to the circumstance that it is a Constitution that we are construing and that the legislative powers that it confers should be construed liberally. There are substantial reasons for thinking that an individual grant of power under the Constitution should be accorded a full operation according to its terms, unrestricted by dubious implications drawn from the existence of another grant of legislative power touching an associated subject matter. There is no inherent reason for supposing that the legislative powers conferred by the Constitution are mutually exclusive; indeed, many instances may be given of overlapping operation. Yet the argument against validity in the present case not only denies this approach to construction but advances to a more extreme conclusion by subtracting from the content of the marriage power, not only what is contained within s. 51(xxii.), but the entire topic of enforcement of the rights, duties and obligations created in the exercise of the marriage power. 1. (1962) 107 C.L.R., at p. 560. 2. (1962) 107 C.L.R., at p. 572. Once this argument is rejected there is no obstacle to the conclusion that, notwithstanding the existence of s. 51(xxii.), the marriage power enables the Parliament to provide for the enforcement of such rights, duties and obligations as may be created in exercise of the marriage power—the view expressed by Quick and Garran, Annotated Constitution of the Australian Commonwealth (1901), p. 608, and adopted by Owen J. in the Marriage Act Case [77] ; Moore, Constitution of the Commonwealth of Australia, 2nd ed. (1910), p. 474, contra; and see Dixon C.J. in the same case [78] . So understood the power may be exercised by providing for the enforcement of rights of maintenance, custody and property by proceedings separate and independent of proceedings for annulment or dissolution of marriage—actual vesting of jurisdiction in specific courts will be accomplished under Ch. III and s. 122. 1. (1962) 107 C.L.R., at p. 602. 2. (1962) 107 C.L.R., at pp. 543-544. Lansell v. Lansell [79] did not decide to the contrary. There, as I have said, s. 86 of the Matrimonial Causes Act 1959 was upheld as a valid law under s. 51(xxii.) because it provided for ancillary relief in a suit for principal relief. That decision did not relate to the marriage power. 1. (1964) 110 C.L.R. 353. It is important now to look more closely at pars (c), (d), (e) and (f) of the statutory definition of "matrimonial cause" as defined by s. 4(1). They are expressed in the following terms: (c) proceedings with respect to— (i) the maintenance of one of the parties to a marriage; (ii) the property of the parties to a marriage or of either of them; or (iii) the custody, guardianship or maintenance of, or access to, a child of a marriage; (d) proceedings between the parties to a marriage for the approval by a court of a maintenance agreement or for the revocation of such an approval or for the registration of a maintenance agreement; (e) proceedings for an order or injunction in circumstances arising out of a marital relationship; or (f) any other proceedings (including proceedings with respect to the enforcement of a decree or the service of process) in relation to concurrent, pending or completed proceedings of a kind referred to in any of paragraphs (a) to (e), including proceedings of such a kind pending at, or completed before, the commencement of this Act. Paragraph (c) opens with the words "proceedings with respect to". Unlike pars (a) and (d) it is not limited to "proceedings between the parties to a marriage". A reference in the Act to a party to a marriage is to be understood as including a reference to a person who was a party to a marriage that has been dissolved or annulled (s. 4(2)). The evident intention to confer a jurisdiction in proceedings not limited to the parties to the marriage is not explained by a desire to accommodate the situation for which s. 4(2) makes provision. Unfortunately by conferring a jurisdiction unlimited as to parties, par. (c) in my opinion travels beyond the marriage power. At least for the same reason pars (e) and (f) of the statutory definition of "matrimonial cause" are also outside the marriage power. Of the four paragraphs in question only par. (d) is a valid exercise of that power. The question then is whether the provisions or any of them can be saved by an application of s. 15A of the Acts Interpretation Act 1901-1973. This is a difficult problem because a reading down may be achieved either by reference to the power contained in s. 51(xxi.) or (xxii.) but the results will differ according to the power which is selected. In R. v. Poole; Ex parte Henry [No. 2] [80] , Dixon J. pointed out that there are two categories of provisions which may be saved by s. 15A. His Honour said: In one type it is found that particular clauses, provisos or qualifications, which are the subject of distinct or separate expression, are beyond the power of the legislature. In the second type, a provision which, in relation to a limited subject matter or territory, or even class of persons, might validly have been enacted, is expressed to apply generally without the appropriate limitation, or to apply to a larger subject matter, territory or class of persons than the power allows. In the first case, the question usually is whether the operation or effect of the remainder of the Act upon the persons or things to which it would apply would be changed if the clauses, provisos and qualifications held bad were excised. In other words, in such a case the right question to ask may be whether liabilities or rights of a different tenor, measure or nature would result. In the second case, the question may simply be whether the legislature intended the provision to have a distributive operation or effect. That is to say, did it intend that the particular command or requirement expressed in the provision should apply to or be fulfilled by each and every person within the class independently of the application of the provision to the others; or were all to go free unless all were bound? 1. (1939) 61 C.L.R. 634, at p. 652. In Pidoto v. Victoria [81] , Latham C.J., after observing that s. 15A should not be read so as to require the court to perform a function essentially "legislative not judicial" by speculating as to an alternative criterion of validity once a provision has been held invalid according to its terms, went on to say [82] : Where the law itself indicates a standard or test which may be applied for the purpose of limiting, and thereby preserving the validity of, the law, the case is different. Thus where a law is clearly made with the intention of exercising the power to make laws with respect to trade and commerce, it is not difficult to read it down so as to limit its application to inter-State and foreign trade and commerce, with which alone the Commonwealth Parliament has power to deal (Constitution, s. 51(i.)). In such a case the subject matter of the legislation itself is such as to provide a test for limiting the law by construction so as to treat it as applying only to that part of a definite subject matter which is within power and with which Parliament clearly intended to deal so far as it could lawfully do so. See also Strickland v. Rocla Concrete Pipes Ltd. [83] , per Gibbs J. 1. (1943) 68 C.L.R. 87. 2. (1943) 68 C.L.R., at pp. 109-110. 3. (1971) 124 C.L.R. 468, at pp. 526-527. Here, as it seems to me, the Parliament by providing for the exercise of a jurisdiction in the matters referred to in pars (c), (d), (e) and (f) independently of proceedings for annulment or dissolution of marriage has sought to exercise primarily the marriage power. For this reason a reading down should be carried out in the first instance by reference to the marriage power with resort to s. 51(xxii.) only to the extent to which the former power is insufficient. Pursuing this approach, I would confine pars (c) (i), (c) (iii) and (e) to proceedings between the parties to a marriage thereby saving par. (c) (i), par. (c) (iii) to the extent to which it relates to the natural and adopted children of the parties to the marriage (cf. s. 5), and par. (e). Paragraph (c) (ii) by reason of its reference to the property of either of the parties to the marriage, presumably comprehending any property howsoever and whensoever acquired, is not susceptible of a reading down under s. 51(xxi.); I would therefore read it down by reference to s. 51(xxii.) and treat it, in conjunction with s. 39, as conferring jurisdiction to grant ancillary relief in proceedings for annulment or dissolution of marriage. I would answer the questions asked by Stephen J. as follows: 1. (i) Yes. (ii) Yes; but only to the same extent as s. 39. (iii) Yes; s. 39 is a valid law of the Commonwealth in so far as the jurisdiction conferred relates to the following paragraphs of the definition of matrimonial cause contained in s. 4: (1) Paragraphs (a), (b) and (d). (2) Paragraphs (c) (i) and (e) but only in so far as they relate to proceedings between the parties to a marriage. (3) Paragraph (c) (ii) but only to the extent to which it relates to proceedings for ancillary relief in relation to proceedings of the kinds referred to in pars (a) and (b) of the definition. (4) Paragraph (c) (iii) but only in so far as it relates to proceedings between the parties to a marriage with respect to the custody, guardianship or maintenance of, or access to, the natural and adopted children of the parties to the marriage. (5) Paragraph (f) but only to such extent as is consequential upon the validity of s. 39 in relation to pars (a), (b), (c), (d) and (e) of the definition of "matrimonial cause". (iv) Yes; but only to the extent to which it relates to proceedings between the parties to a marriage with respect to the natural and adopted children of the parties to the marriage. (v) Yes; in so far as it relates to proceedings between the parties to a marriage or between the parties to a marriage, or one of them, and the natural or adopted child of that marriage. (vi) Yes; but only in so far as such proceedings are brought to obtain relief as an ancillary to proceedings of the kinds referred to in pars (a) and (b) of the definition of "matrimonial cause". 2. Not necessary to answer. 3. (a) No. (b) No. (c) No. Jacobs J. There are two matters removed into this Court pursuant to s. 40A of the Judiciary Act 1903-1973 Cth. The first, Russell v. Russell, comes from the Supreme Court of Victoria. It is a proceeding for dissolution of marriage and for orders for custody of children, maintenance of the petitioner and the children and settlement of property. It was instituted under the Matrimonial Causes Act 1959-1966 Cth but after the Family Law Act 1975 Cth came into force a notice under s. 9(2) was filed on behalf of the respondent. Section 9(2) provides that: 9. (2) Where the parties have lived separately and apart for a continuous period of not less than 12 months immediately preceding the date of commencement of this Act, pending proceedings for a decree of dissolution of marriage shall, if the applicant so requests, be dealt with as if they were proceedings instituted under this Act on the ground referred to in section 48, and in relation to proceedings in which such a request is made, sub-section 48 (2) has effect as if the proceedings for dissolution of marriage had been instituted by an application filed on the date of commencement of this Act. When the matter came on for hearing in the Supreme Court the question was raised whether in view of s. 97 of the Family Law Act 1975 Cth it should be heard in closed court and with Judge and counsel unrobed and this led to the question whether s. 97 was within Commonwealth power and to further questions on the validity of the Family Law Act 1975 Cth. Kaye J. determined to proceed no further with the matter but to regard it as removed into the High Court under s. 40A. On its removal it came before Stephen J. who on 20th February 1976 reserved for the Full Court pursuant to s. 18 of the Judiciary Act 1903-1973 certain questions in respect of the validity and effect of the Family Law Act 1975. Those questions are as follows: 1. Is (i) section 97 in so far as it deals with courts other than the Family Court (ii) section 9(4) (iii) section 39 (iv) section 64 (v) section 74 (vi) section 78 of the Family Law Act 1975 a law of the Commonwealth validly enacted pursuant to the powers conferred by pars (xxi.), (xxii.) and (xxxix.) of s. 51, s. 77(iii.) or any other provision of the Constitution? 2. Is s. 123 of the said Act to the extent that it authorizes the making of regulations for or in relation to the practice and procedure to be followed in Courts other than the Family Court exercising jurisdiction under the Family Law Act 1975 beyond the power of the Parliament of the Commonwealth to any and if so what extent? 3. In so far as s. 97 of the said Act is a valid law of the Commonwealth does non-compliance with the provisions of (a) sub-section (1) (b) sub-section (3) (c) sub-section (4) thereof by a Court (other than the Family Court) exercising or purporting to exercise jurisdiction under the Act affect the validity of any decree or order of such Court made in exercise or purported exercise of such jurisdiction? The second matter, Farrelly v. Farrelly, comes from the Supreme Court of South Australia. It is an application for custody of the child of a marriage made to the Supreme Court of South Australia under the provisions of the Guardianship of Infants Act, 1940 S.A. by originating summons issued on 10th October 1975. The proceedings were thus commenced before the Family Law Act 1975 came into force. When the matter came on for further hearing of an interlocutory application on 21st January 1976 a question arose whether the proceedings being pending proceedings constituting a matrimonial cause within the meaning of those words as defined in s. 4(1), not being proceedings for principal relief, and having been instituted under the law of South Australia, should be dealt with as if they were proceedings instituted under the Family Law Act 1975. By s. 9(4) of that Act they were required so to be dealt with. But a question arose in the mind of the judge whether the Family Law Act 1975 was a valid exercise of Commonwealth power, at least in so far as it applied to a proceeding with respect to the custody, guardianship or maintenance of, or access to, a child of a marriage subsisting between parties to proceedings who were not seeking dissolution of their marriage or other principal relief. Upon removal the matter has been heard together with the first matter. I shall deal first with the question raised in this second matter. Part VII of the Family Law Act 1975 deals with the welfare and custody of children of a marriage. Part VIII deals inter alia with the maintenance of such children. By s. 5 the words "a child of the marriage" include, inter alios, a child of either the husband or the wife if at the relevant time the child was ordinarily a member of the household of the husband and wife. The "relevant time" is defined in s. 5(2). Part VII is not limited to questions of custody and welfare arising when proceedings have been instituted for "principal relief", that is to say, for dissolution of marriage, nullity of marriage or declaration of validity of marriage or of its dissolution or nullity or arising when decrees in such proceedings have been made. The Act purports to deal with the law governing custody and welfare of a child of a marriage generally and with the powers and duties of a court in proceedings with respect thereto. Nor is Pt VII limited to proceedings between the parties to a marriage. Both it and the definition of matrimonial cause in s. 4 and the extent of the jurisdiction in matrimonial causes conferred by Pt V of the Act manifest an intention that a court exercising jurisdiction under the Act may make an order for the custody of a child of a marriage at the suit both of a party to the marriage and of a third party and whether or not the marriage has terminated by death of one or both of the parties thereto or by dissolution thereof. Likewise Pt VIII of the Act is intended to apply to maintenance of a child of a marriage when it is sought in proceedings where no principal relief is claimed, and when the marriage has terminated by the death of one of the parties thereto (ss. 73, 74, 82). Subject to an exception found in s. 79 to which I hereafter refer, in both Pt VII and Pt VIII the power of the Court to make orders is made to depend upon the child being below a certain age. Section 61(2) provides that an order with respect to the custody or guardianship of, or access to, a child shall not be made in respect of a child who has attained the age of eighteen years. Section 73 provides substantively that the parties to a marriage are liable, according to their respective financial resources, to maintain the children of the marriage who have not attained the age of eighteen years. By s. 51 of the Constitution the Parliament of the Commonwealth may make laws relating to (xxi.) Marriage; (xxii.) Divorce and matrimonial causes; and in relation thereto, parental rights, and the custody and guardianship of infants. Paragraphs (xxi.) and (xxii.) of s. 51 are the only subject matters of Commonwealth power which are not related to what may be broadly described as public economic or financial subjects but which are related to what are commonly thought of as private or personal rights. The reason for their inclusion appears to me to be two-fold. First, although marriage and the dissolution thereof are in many ways a personal matter of the parties, social history tells us that the state has always regarded them as matters of public concern. Secondly, and perhaps more importantly, the need was recognized for a uniformity in legislation on these subject matters throughout the Commonwealth. In a single community throughout which intercourse was to be absolutely free provision was required whereby there could be uniformity in the laws governing the relationship of marriage and the consequences of that relationship as well as the dissolution thereof. Differences between the States in the laws governing the status and the relationship of married persons could be socially divisive to the harm of the new community which was being created. But because these subject matters enter so deeply into the field of personal and private rights, a field largely left within the legislative power of the States, it is difficult to resist the tendency in oneself to regard the Commonwealth power as an intrusion into an area solidly filled by State laws on a variety of subject matters which govern personal and private rights. The tendency to minimize the unique intrusion by a narrow characterization of the subject matters so that as far as possible the framework of State laws governing personal and private rights is preserved must be resisted. One construction might be that s. 51(xxi.) is a power limited to prescribing the formal and substantive requirements for a valid marriage (cf. Harrison Moore, Constitution of the Commonwealth of Australia, 2nd ed. (1910), pp. 474-475) and that s. 51 (xxii.) is limited to divorce and to that limited class of legal proceedings which at the date of the establishment of the Commonwealth were commonly heard and determined before the same tribunal as dealt with legal proceedings for divorce; and that in relation to such legal proceedings and not otherwise there is power to make laws relating to parental rights and the custody and guardianship of infants. However it seems to me that there a number of difficulties in such a limited construction. It is established by Attorney-General (Vict.) v. The Commonwealth [1] that the power to make laws relating to marriage is not limited to laws relating to the formal and substantive requirements of a valid marriage. Legitimation of children by subsequent marriage of their parents was held to fall within the marriage power. A provision that a child shall be deemed for all purposes to be legitimate where the marriage was void and certain other circumstances existed was also held to fall within the marriage power. The word "marriage" is used in the Constitution not only to describe the act ceremony or process of marriage but the institution, the relationship between man and woman which results from the act ceremony or process. Taylor J. said [2] : the word "marriage"—appearing without limitation or qualification—is entitled to as wide an interpretation as it can reasonably bear. It is, of course, impossible to suggest a synonym which will precisely define the limits of the power and, no doubt, its full meaning will be worked out only in the fullness of time. But in the meantime I feel bound to regard the paragraph as justification for the enactment of any law with respect to marriage considered as an institution. That is to say, that it extends not only to laws prescribing the form and requisites of a valid marriage but also to laws defining and regulating the respective rights duties and obligations of the parties inter se. Indeed, the full measure of the legal effect of a marriage can be determined only by reference to the rights duties and obligations, which, by law, arise out of the relationship and I can see no reason why a constitutional power to make laws with respect to the subject matter of "Marriage" should not be thought to authorize laws defining or modifying and re-defining the legal incidents of the relationship. 1. (1962) 107 C.L.R. 529. 2. (1962) 107 C.L.R., at p. 560. Once the word "marriage" is given a meaning which goes beyond the act, ceremony or process by which the legal relationship of husband and wife is constituted and includes the institution itself, the nature and basis of that institution must be examined in order to determine the extent of a power to make laws relating thereto. The act of marriage is designed to create rights and duties between the parties, some of which may be of immediate legal operation and some of which may be recognized but not enforced by the law. The enunciation, recognition and enforcement of matrimonial rights and duties are part of the subject matter of marriage regarded as an institution and not merely as a ceremony. Further, marriage as a social institution which the law clothes with rights and duties attaching to the parties thereto is primarily an institution of the family. It is true that marriage can be regarded as a social relationship for the mutual society help and comfort of the spouses but it cannot be simply so regarded. The primary reason for its evolution as a social institution, at least in Western society, is in order that children begotten of the husband and born of the wife will be recognized by society as the family of that husband and wife. Marriage would appear to be a product of the recognition actual or instinctive of participation by the male in the process of conception, its purpose being in order that the male may and will participate in the nurture and protection of the offspring. [T]he establishment of marriage in all civilized states is built on this natural obligation of the father to provide for his children: for that ascertains and makes known the person who is bound to fulfill this obligation: Blackstone's Commentaries, 15th ed. (1809), vol. 1, p. 447. The recognition by society of rights and duties of husband and wife in respect of the children of their marriage and of the relationship of the children of that marriage to their parents springing from their status as children of the marriage lies not on the periphery but at the centre of the social institution of marriage. In societies or stages in the development of any society where concubinage has been legally recognized, a main distinction therein from marriage has been in the status of the children of such a union with consequent differences in the obligations of the father to the children. In those systems of slavery best known to us—the Roman and the United States—the marriage of slaves was not recognized during the continuance of their bondage in so far as the children of a union of slaves were not recognized as part of the family of the two slaves; but on emancipation the matrimony took effect and the children were so recognized without further ceremony. "Marriage is rooted in the family" wrote Westermarck in his History of Human Marriage. There can be discerned in modern times a concept of the family being rooted in marriage, the so-called "nuclear family"; but this is not the concept which has underlaid the institution of marriage in our society and in our law. The procreation of children and the nurture of them is not a mere consequence of marriage, whereby the children may be regarded as collateral to the real subject matter of marriage. If a government were to forbid marriage, it would be forbidding not the procreation of children but the nurture of children in a close relationship to their parents recognized and ordered by society. A forbidding of marriage but a recognition that a man and a woman could cohabit and have children to whom they as parents had a close recognized and ordered relationship would be to forbid the ceremony of marriage and nothing more. The nurture of children by, and in recognized and ordered relationship with, their parents is thus integral to the concept of marriage as it has developed as an institution in our society. Laws relating to the nurture of children of a marriage are laws relating to marriage as a social institution. Custody and guardianship and maintenance of children of a marriage are aspects of the nurture of children within the marriage relationship. What is argued to the contrary is that laws relating to custody, guardianship and maintenance of the children of a marriage are laws relating to children and not to marriage. The distinction thus sought to be made creates a false dichotomy. Such laws do indeed relate to children but the law relating to children of a marriage is a law particularly applicable because the children are children of a marriage. It is a branch of the law governing the relationship of the spouses between themselves and between them and their children begotten of their union. The distinctness of the laws governing children of a marriage has tended to become obscured by statute law of the past century designed to ameliorate the position of children generally, including children of a marriage and more recently to assimilate the position of children born outside marriage with the position of children of a marriage. However, the distinctness in subject matter must still be observed when the question being considered is the extent of constitutional power. The argument that par. (xxi.) is not wide enough to comprehend laws relating to the children of a marriage largely depends upon the expression in par. (xxii.) of the power to make laws relating to parental rights and the custody and guardianship of infants in relation to divorce and matrimonial causes. The expression of this power in the one case, it is said, implies their exclusion in the case of the marriage power. It is a fragile argument when the subject is the extent of constitutional power, but it is baseless. The expression of the power in par. (xxii.) can be explained by the fact that it might have been arguable that even though it was clear that so long as a marriage subsisted laws relating to custody, guardianship and maintenance of children fell within the marriage power, when a marriage was dissolved the marriage power was no longer sufficient to support such laws and that, despite the current legislative practice of the colonies in 1900, a divorce and matrimonial causes power simpliciter would not include the matter of custody, guardianship and maintenance of children of the dissolved marriage. I conclude therefore that the power to make laws relating to marriage includes a power to make laws relating to the nurture of, and consequently the custody, guardianship and maintenance of, a child of the marriage. My reasons for this conclusion do not depend on the continuance of the marriage but on a recognition that by marriage and the procreation of children in the marriage relationship each parent has social rights and duties of nurture in respect of those children which arise from and are part of the marriage relationship which exists or which previously existed. The power to make laws giving the custody or guardianship of a child of a marriage to a third party is ancillary to the power to make laws for the nurture of children of a marriage by the parties to the marriage. Where the parties to a marriage, or the survivor of them, fails or fail in the duty of nurture the rights which a law gives to them in respect of guardianship and custody may be taken away and given to another. Where the parties to a marriage are dead, a child of a marriage may still need the nurture which the parties to the marriage would or should have given him or her and it is ancillary to the power to make laws for the nurture of a child of a marriage by its parents to make laws for the care of a child of a marriage who is deprived by death of that care. I would remit Farrelly v. Farrelly to the Supreme Court of South Australia to be dealt with under the provisions of the Family Law Act 1975. I turn now to the matter removed from the Supreme Court of Victoria. The questions reserved by Stephen J. fall into two main categories. The first category requires the determination of the question whether the jurisdiction conferred under the Family Law Act 1975 extends to the determination of matters outside the legislative power of the Commonwealth, and, if so, the consequential effect on validity of certain sections of the Act. Part V of the Family Law Act 1975 deals with jurisdiction in matrimonial causes. Section 39(1) which is in Pt V provides that subject to that Part, a person may institute a matrimonial cause under the Act in the Family Court (created under the Act) or in the Supreme Court of a State or a Territory. By s. 39(2), again subject to Pt V, a person may institute a matrimonial cause, not being proceedings for principal relief, in a court of summary jurisdiction of a State or Territory. By s. 39(5) the Supreme Court of each State is invested with federal jurisdiction to hear and determine matrimonial causes instituted under the Act and matrimonial causes continued in accordance with s. 9. Sub-section (4) of the latter section is referred to by me earlier in these reasons. "Matrimonial cause" is defined in s. 4(1) to mean (a) proceedings between the parties to a marriage for a decree of— (i) dissolution of marriage; or (ii) nullity of marriage; (b) proceedings for a declaration as to the validity of a marriage or of the dissolution or annulment of a marriage by decree or otherwise; (c) proceedings with respect to— (i) the maintenance of one of the parties to a marriage; (ii) the property of the parties to a marriage or of either of them; or (iii) the custody, guardianship or maintenance of, or access to, a child of a marriage; (d) proceedings between the parties to a marriage for the approval by a court of a maintenance agreement or for the revocation of such an approval or for the registration of a maintenance agreement; (e) proceedings for an order or injunction in circumstances arising out of a marital relationship; or (f) any other proceedings (including proceedings with respect to the enforcement of a decree or the service of process) in relation to concurrent, pending or completed proceedings of a kind referred to in any of paragraphs (a) to (e), including proceedings of such a kind pending at, or completed before, the commencement of this Act; It may be that "matrimonial cause", as defined in the Act, is wider than "matrimonial cause" within the meaning of par. (xxii.) of the Constitution, but there is nothing to prevent the legislature giving the words a wider meaning in the Act, provided that the subject matter of the wider meaning is within Commonwealth power. The parts of the definition of "matrimonial cause" which are claimed to be beyond power when read with s. 39 are pars (c), (d) and (e) and that part of (f) which refers to proceedings of the kind referred to in those paragraphs. Paragraph (c) (i) and (iii) and pars (d) and (e) are within the power contained in s. 51(xxi.) of the Constitution for the reasons which I have earlier expressed in respect of the matter of Farrelly v. Farrelly. For the same reasons ss. 64 and 74 of the Act are valid. Paragraph (c) (ii) of the definition of "matrimonial cause" requires particular attention as a conclusion upon it is not wholly covered by my earlier reasons. If the words are read in isolation they cover proceedings unrelated to the marriage relationship. The example given during argument was that of an action for specific performance of an agreement for sale of land made between parties who were not but had previously been married (see the meaning given to "party to a marriage" by s. 4(2)). However, although the definition section does not define the various words expressly "for the purposes of the Act" it is clearly intended that that should be so. The proceedings referred to in par. (c) (ii) are proceedings not only instituted under s. 39 but to which the substantive provisions of the Act apply. Those provisions are found in ss. 78 and 79 and, in respect of maintenance, ss. 74 and 80. If these provisions (or any of them) are within power then a content of par. (c) (ii) of the definition of "matrimonial cause" is within power. Sections 74 and 80 are covered by my earlier reasons. The purpose and operation of s. 79 is clear. Even if in law or equity the title to property is in one party to a marriage, the circumstances of the marital relationship may make an alteration of property interests just and equitable. In the case of the parties to a marriage, the marriage relationship brings this provision within the marriage power. The position is quite different from a legislative attempt to seize upon the bare fact of marriage as a ground for altering proprietary rights unrelated to the marriage relationship. Section 79(4) relates the power to the subject matter of the marriage relationship and s. 79(3) if this be necessary (see Lansell v. Lansell [3] ) limits the occasions when an order can be made. 1. (1964) 110 C.L.R. 353, at p. 359. A different situation arises in respect of the application of s. 79 to a child of the marriage. The section in this respect would apply to a child of any age and its application might be unrelated to rights and duties arising from the institution of marriage. It is unnecessary to express a concluded view upon the validity of the inclusion in the section of the references to a child of the marriage, but, if they be beyond power, they are clearly severable. Also, the validity of s. 78 does not arise for decision in the instant case and I would refrain from expressing a view thereon until the particular facts of a particular case fall to be considered. The provision is severable and does not affect the validity of other sections of the Act. The proceedings between the parties in Russell v. Russell do not claim any relief which requires an application of the power purported to be conferred on a court by s. 78. The second category of question referred by Stephen J. requires a determination of the question whether the Commonwealth, on investing State courts with federal jurisdiction to determine matters under the Family Law Act 1975, had power to make laws relating to the practice and procedure of those courts in the exercise of that jurisdiction; and, if so, whether s. 97 is a law relating to practice and procedure or a law which goes further and alters the constitution of the State court. Section 97 is as follows: 81. (1) Subject to sub-section (2) and to the regulations, all proceedings in the Family Court, or in another court when exercising jurisdiction under this Act, shall be heard in closed court. (2) Subject to the regulations, relatives or friends of either party, marriage counsellors, welfare officers and legal practitioners may be present in court unless in a particular case the court otherwise orders. (3) In proceedings under this Act, the court shall proceed without undue formality and shall endeavour to ensure that the proceedings are not protracted. (4) Neither the Judge hearing proceedings under this Act nor counsel shall robe. Associated with the question of the validity of s. 97 is the question whether s. 123 of the Act to the extent that it authorizes the making of regulations for or in relation to the practice and procedure to be followed in courts other than the Family Court exercising jurisdiction under the Family Law Act 1975 is beyond the power of the Parliament of the Commonwealth to any and if so what extent. The submission that these sections to the extent indicated are beyond power depends upon the doctrine that when investing a State court with federal jurisdiction under s. 77(iii.) of the Constitution, the Parliament cannot thereby alter the constitution and organizational structure of a State court. Le Mesurier v. Connor [4] ; Adams v. Chas. S. Watson Pty. Ltd. [5] ; Peacock v. Newtown Marrickille and General Co-operative Building Society No. 4 Ltd. [6] ; Kotsis v. Kotsis [7] . Therefore the Commonwealth cannot make a Commonwealth officer a functionary of a State court: Le Mesurier v. Connor [4] . Nor can it invest an officer of a State court with judicial power under the Constitution: Kotsis v. Kotsis [7] . However, if the structure of the court remains intact, and the function only is altered, the provision cannot be impugned. See per Latham C.J. in Adams v. Chas. S. Watson Pty. Ltd. [8] . However, the Parliament in its investing of a State court with federal jurisdiction is not limited by the existing jurisdiction of the State court in any sense of the word "jurisdiction". The investing with federal jurisdiction will in one sense itself be an extension of any existing jurisdiction of a State court. If the word "jurisdiction" is used in the sense of the extent or limitation thereof by reference to subject matter or the value thereof, place or time for commencement of proceedings or the like, the Commonwealth Parliament is not limited by provisions in State law in these respects, though by s. 39(2) of the Judiciary Act 1903-1973 it has recognized and respected some of these limitations on jurisdiction. The Family Law Act 1975 prescribes its own limitations on the jurisdiction which can be exercised by particular State courts, e.g. s. 39. 1. (1929) 42 C.L.R. 481. 2. (1938) 60 C.L.R. 545. 3. (1943) 67 C.L.R. 25. 4. (1970) 122 C.L.R. 69. 5. (1929) 42 C.L.R. 481. 6. (1970) 122 C.L.R. 69. 7. (1938) 60 C.L.R., at p. 555. There is nothing in the doctrine enunciated in Le Mesurier v. Connor [4] , and acknowledged in the other cases to which I have referred, which would support the view that the Commonwealth Parliament when it invests a State court with Federal jurisdiction cannot prescribe the practice and procedure to be followed by parties who invoke, and by the Court which exercises, the invested jurisdiction. If the laws on the practice and procedure are within the subject matter of Commonwealth power and do not alter the structure of the State court then they offend no established principle. Section 97 is within the subject matter of Commonwealth power contained in s. 51(xxi.). Section 97(1) operates to ensure that in matters concerning the marriage relationship and the children of the marriage there shall be a large degree of privacy. One is tempted to refer to analogous legislative provisions in this field which recognize the desirability of some such privacy but to do so would be to fall into the error of testing the extent of Commonwealth power in respect of particular legislation by the novelty or lack of novelity in its concept. However, in another field an example may be taken. In an action for infringement of a patent heard in a State court exercising federal jurisdiction it would surely be open to the Commonwealth to legislate that evidence of the alleged infringing process if it be claimed to be a secret process be heard in a closed court. I cannot see why such a provision being within the subject matter of power should nevertheless be invalid as a matter affecting the constitution or organizational structure of the State court. Similarly in the present case the purpose and effect of the section as a whole being the avoidance of formality and publicity in proceedings under the Act, the avoidance thereof is within the subject matter of power. If then, being otherwise within power, do any of the provisions of s. 97 alter the constitution or organizational structure of the Supreme Court of Victoria? They do not. The same Court sits whether the place of hearing be open or closed to the public and whether or not any particular form of dress is prescribed. The objections which have been raised seem to me to depend ultimately on a denial that such prescriptions as are found in s. 97 are within the subject matter of the Commonwealth power itself rather than on the assertion that the structure of the court is altered. But this objection is met by the special nature of the proceedings under the Family Law Act 1975. 1. (1929) 42 C.L.R. 481. Lastly I would state my opinion that the provisions of s. 97 are directory only and that a failure to comply with them does not invalidate the proceedings in which the failure occurs. I would therefore answer the referred questions as follows: 1. (i) Yes. (ii) Yes. (iii) Yes. (iv) Yes. (v) Yes. (vi) Not answered. 2. No. 3. (a) No. (b) No. (c) No. I would remit the matter to the Supreme Court of Victoria for further hearing and determination.