BAILII [Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]

The Law Commission

You are here: BAILII >> Databases >> The Law Commission >> Blood Tests and the Proof of Paternity in Civil Proceedings (Report) [1968] EWLC 16 (01 January 1968)
Cite as: [1968] EWLC 16

[New search] [View without highlighting] [Printable PDF version] [Help]


    The Law Commission

    (LAW COM. No. 16)

    Laid before Parliament by the Lord High Chancellor pursuant to section 3(2) of the Law Commissions Act 1965
    Ordered by The House of Commons to be printed 30th October 1968
    The Law Commission was set up by section 1 of the Law Commissions Act 1965 for the purpose of promoting the reform of the law. The Commissioners are—
    The Honourable Mr. Justice Scarman, O.B.E., Chairman. Mr. L. C. B. Gower, M.B.E. Mr. Neil Lawson, Q.C. Mr. N. S. Marsh, Q.C. Mr. Andrew Martin, Q.C.
    Mr. Arthur Stapleton Cotton is a special consultant to the Commission. The Secretary of the Commission is Mr. J. M. Cartwright Sharp, and its offices are at Lacon House, Theobald's Road, London, W.C.I.

    A. INTRODUCTION....... 1
        (a) Affiliation Proceedings ..... 9
        (b) Divorce and Nullity Proceedings 11
        (a) The Presumption of Legitimacy .... 31
        (b) Blood Tests undertaken voluntarily . . * . 32
        (c) Power to direct the use of blood tests 33
        (d) Joint defendants ....... 35
        (e) Refusal to be tested ...... 38
        (f) Refusal by a child or mentally disordered person 48
        (g) Non-exclusion results as evidence .... 50
        (h) Procedure in affiliation proceedings 53
        (i) Proceedings other than affiliation .... 57
        (j) Testing procedure . . 58
    Appendix A—Draft Clauses . . . . . A
    Appendix B—The Nature of Blood Group Evidence . B
    Appendix C—Organisations consulted .... C

    LAW COMMISSION Item XIX of the Second Programme
    To the Right Honourable the Lord Gardiner, the Lord High Chancellor of Great Britain
  1. At the beginning of 1966 the Judges of the Probate, Divorce and Admiralty Division of the High Court suggested that the Law Commission might find it appropriate to consider the whole question of blood tests in paternity issues and, in particular, whether or not it would be right to amend the law to enable the courts to order parties to undergo blood tests. We accordingly considered this suggestion as we are bound to do under s. 3(1) of the Law Commissions Act 1965, and in July 1967 we published a Working Paper,[1] which was given a wide circulation, setting out our preliminary views on the subject.
  2. As is our practice, we consulted a wide range of organisations representing the legal profession, teachers of law, the medical profession and other sections of the community. In addition we consulted Government departments, the Ministries of Justice of three countries whose courts have the power to order compulsory blood tests, and many individuals both in this country and abroad. This subject clearly involves many questions which we, as lawyers, would not have been qualified to answer without the guidance and views of those whom we consulted.[2] Our debt to them is enormous; we wish to record our appreciation and gratitude for the advice which was given to us. In particular we wish to record our debt to the Home Office who, throughout our work on this topic, have given us invaluable help. Although, naturally, the views which we received conflicted to some extent it is a significant fact, which we feel should be mentioned, that we did not receive one opinion which disagreed with the central proposition in our Working Paper, that the courts should have the power to order compulsory blood tests in certain circumstances.
  3. We decided to confine our enquiry to civil proceedings since the powers of the court to obtain evidence in criminal proceedings raise some quite distinct issues. Still less are we here concerned with blood tests carried out to determine the level of alcohol in the blood stream or the presence of drugs or other similar substances. Thus limited to the determination of paternity, the subject of blood tests falls within the field of family law, and this study forms part of our general review of family law under Item XIX of our Second Programme. A preliminary consideration of the issues made it clear that we should not examine blood tests in isolation but should also consider whether any of the existing rules of evidence would need to be altered. While, therefore, this Report deals predominantly with the advisability of giving the court the power to order blood tests, we have considered this in conjunction with the wider question: is the law governing the proof of paternity satisfactory?
  4. In our Working Paper we suggested that the privilege from self-incrimination, which entitles a party or witness to refuse to answer a question tending to prove that he or she has committed adultery, should be abolished. We also argued that the ordering of blood tests would not offend the principle underlying this privilege, even if it were to be retained. Since our Working Paper was circulated the Law Reform Committee have recommended that the privileges relating to adultery, embodied in ss. 43(1) and (2) of the Matrimonial Causes Act 1965, should be abolished.[3] This will be done by the Civil Evidence Bill which is expected to become law before the end of the 1967/68 Session.
  5. We attempt to summarize the information which we have received regarding the scientific basis of blood group evidence and the possibility of errors in blood grouping in Appendix B.[4] We are satisfied that as medical knowledge stands at present blood tests may provide conclusive evidence in a negative sense; that is, they can prove that a given man could not, according to the biological laws of heredity, be the father of a particular child. They cannot prove conclusively that he is the child's father but they can show, with varying degrees of probability, that he could be. Where blood tests indicate that the man concerned could not be the child's father we shall term this an "exclusion result"; where the tests indicate that he could be the father we shall term this a "non-exclusion result." Where a man is wrongly accused of paternity there is now at least a 70 per cent chance that an exclusion result will be obtained from blood tests. The chances of obtaining an exclusion result will no doubt be increased as further blood groups are discovered and more refined techniques developed. Even now, where uncommon blood factors are present or where, for example, it is known that the father must be one of only two men both of whose blood groups are known, the chances of excluding a wrongly accused man can be very much greater than 70 per cent.[5] Where blood tests provide a non-exclusion result they indicate a possibility that the man concerned is the child's father. The strength of this indication will depend primarily upon the incidence of the relevant blood factors in the population. Where common blood factors are present there may be a statistical possibility that any one of, say, 50 per cent of the adult male population could be the child's father, but in an extreme case where uncommon blood factors are present the incidence of possible fathers could be as low as one in fifty million. The circumstances of an individual case may also increase the evidential value of a non-exclusion result. The presence of certain factors in a child's blood, for example, may be almost conclusive proof that a parent is of a particular ethnic group. There may be cases, for example the Scottish case of Sinclair v. Rankin[6] where it is clear that the father of a child must be one of only two men. In such a case a non-exclusion result in respect of blood tests on the first man and an exclusion result in respect of the second man might be acceptable as proof that the first man is the child's father.
  6. Important evidence bearing on paternity may also be provided by anthropological tests. These tests involve a physical examination of the parties concerned and an analysis of various characteristics which are known to be inherited according to the accepted laws of genetics. We understand, for example, that the possibility of a child with brown eyes being born to a man and woman both of whom have blue eyes can be calculated statistically. Similarly, these tests may reveal the presence of webbed-toes or some other physical peculiarity in both the child and putative father, the importance of which in determining the child's paternity can, it is said, be expressed statistically. Anthropological tests are used, for example, by the courts of Denmark which may order them when there are two (or more) possible fathers, neither of whom is excluded by blood tests. There seems to be some doubt within the medical profession in this country as to the accuracy of anthropological tests and we understand that the necessary statistical material has not been compiled here. We do not recommend therefore, at this stage, that the courts should be given the power to order anthropological examinations. However, the courts of several continental countries do use these tests and we would hope that they will be scientifically evaluated by geneticists and the medical profession. There appears to be nothing to prevent anthropological evidence, as an indication of paternity, being submitted in any proceedings where paternity is in issue but equally there is no authority that we are aware of allowing courts to order anthropological examinations.
  7. The main types of civil proceedings in which we contemplate blood tests playing an important part are those where paternity is an issue in suits for divorce or nullity, petitions for legitimacy declarations and affiliation proceedings. We also bear in mind the usefulness of blood tests in determining succession rights.[7] Paragraph 39 of the Report of the Committee on the Law of Succession in relation to Illegitimate Persons,[8] under the chairmanship of Lord Justice Russell, draws attention to the increased value that would be attached to affiliation orders as evidence in proceedings concerning succession rights, if blood tests could be ordered. Even where there has been no affiliation order, blood tests could clearly be of assistance as evidence in any proceedings where an illegitimate person claims an inheritance through living persons or through a deceased person whose blood grouping is known.
  9. Where paternity is in issue, the court has to arrive at a finding of fact, namely, whether a certain man is or is not the father of a certain child. In considering the part which blood tests at present play in assisting the court to arrive at this decision we have examined two separate questions:
  10. (a) What is the nature of the evidence required by the court to prove or disprove paternity? and
    (b) Does the court have power as the law stands at present to order blood tests?
    (I) The Nature of the Evidence Required
    (a) Affiliation Proceedings
  11. Affiliation proceedings[9] clearly comprise the great majority of cases in which paternity is in issue. These proceedings enable the mother of an illegitimate child to obtain payment for its maintenance from the father. In recent years there has been a marked rise in the number of illegitimate births and recent
  12. " statistics show that this is a continuing increase.[10] The number of applications for affiliation orders (under 10,000 per annum) is small in comparison with the total number of illegitimate births (67,000 in 1966). No doubt the great majority of these children are born into stable unions or are adopted soon after birth. Even where the child is not born into a stable union or adopted the father may voluntarily enter into a maintenance agreement without the mother having first to institute affiliation proceedings. Nevertheless, in spite of the well-known reluctance of women to institute this type of proceeding, there were 8,664 applications (7,458 orders) in 1966 and 9,634 applications (8,507 orders) in 1967. It is not possible to establish in how many of these cases the court might have been assisted if it could have ordered blood tests, but it is widely accepted that the court often has extreme difficulty in arriving at a decision on the evidence available under the present law.
  13. In affiliation proceedings the mother[11] (the complainant) must prove that the defendant is the father of her child but she is only required to prove this on a balance of probabilities and not beyond all reasonable doubt. On the other hand the evidence given by the complainant as to the paternity of her child must be corroborated[12] in some "material particular."[13] She must produce independent evidence implicating the defendant, but this evidence may be direct or circumstantial, though it must be such that it shows more than a possibility that the defendant is the father of the complainant's child. For example, where there is evidence that over a long period (including the time of conception) the defendant associated with the complainant on terms of affection and there is no evidence that the complainant associated with other men, the court may treat such evidence as sufficient corroboration of the complainant's evidence.[14]
  14. (b) Divorce and Nullity Proceedings
  15. In divorce and nullity proceedings, unlike affiliation proceedings, the decision of the court is under the constraint of the presumption of legitimacy.[15]This long standing rule of English Law was authoritatively stated by the Lord Chief Justice of the Court of Common Pleas delivering the unanimous opinion of the Judges in the Banbury Peerage Case :[16]
  16. "the birth of a child from a woman united to a man by lawful wedlock is, generally, by the law of England, prima facie evidence that such a child is legitimate."

    The strength of this presumption has varied at different periods. At one time it could be rebutted only by showing that the husband could not be the father because he was "beyond the four seas" during the whole of the possible period of conception,[17] but in modern times it may be rebutted by satisfactory evidence that the husband did not have sexual intercourse with his wife at any relevant time. But even this was made extremely difficult to prove as a result of what came to be known as the rule in Russell v. Russell.[18]This rule, which prohibited both the husband and wife from giving evidence of "non-access" at the time of the child's conception, was, however, abolished by the Law Reform (Miscellaneous Provisions) Act 1949. Nevertheless it is still not sufficient for the husband to show that he was probably not the father; he must prove beyond reasonable doubt that he was not. For example, the fact that the wife can be shown to have committed adultery with any number of men is not by itself enough to rebut the presumption, for it does not exclude the possibility of the husband also having had intercourse with her.[19] It was because the husband in Watson v. Watson[20] could not demonstrate beyond all reasonable doubt that the child was not his that he was adjudged by Barnard J. to be liable for its maintenance, in spite of the fact that the wife had refused a request made by the husband for tests to be made on the child's blood. The judge concluded[21] that the child was probably not the husband's but that he was constrained by the presumption of legitimacy to decide in favour of the wife.

  17. Recently Lord Denning has said,[22] in the Court of Appeal, that the presumption of legitimacy is now rebuttable on a balance of probabilities rather than on proof beyond all reasonable doubt. He held that, as the burden of proof required to prove adultery is, since Blyth v. Blyth,[23] the same as that in other civil proceedings (i.e. on a balance of probabilities), by analogy the criminal standard of proof beyond all reasonable doubt could no longer apply to the presumption of legitimacy. We shall discuss Blyth v. Blyth later but in our view there is no doubt that it cannot be regarded as authority for Lord Denning's proposition in ReL; his remarks about the presumption of legitimacy were obiter and Willmer L.J., supported expressly by Arthian Davies L.J., said that he was not "prepared to hold that the presumption of legitimacy can be rebutted on a mere balance of probabilities."[24]
  18. Historically it is easy to see why the standard of proof required of a complainant in affiliation proceedings to establish paternity is lower than that required in matrimonial proceedings to rebut the presumption of legitimacy. In affiliation proceedings there is normally no argument about whether the child is illegitimate; the issue is: who is the father? Where, however, an attempt is made in matrimonial proceedings to rebut the presumption of legitimacy the attempt, if successful, will bastardise a child who would otherwise be legitimate. In former times bastardy was a source of reproach and ridicule and the strength of the presumption of legitimacy was a reflection of this attitude. Furthermore the financial prospects of an illegitimate child could be bleak. Today, society's views on illegitimacy have moderated and the child is not placed under such grave material disadvantages; for example, supplementary benefits are now payable to the mother as of right. However, until the Russell Committee Report[25] is implemented there will remain important differences between the status of legitimate and illegitimate children, notably in their rights of succession.
  19. Despite these changes in public and legal attitudes towards illegitimacy, the courts naturally still regard it as a very serious matter to make a decree which has the effect of bastardising a child.[26] However, we suggest that in most cases it is in a child's best interests to know, if possible, the true position as to its paternity. Where a husband has denied being the father of his wife's child, but has been unable because of the strength of the presumption of legitimacy to prove that he is not, the emotional and financial effect on the child is not likely to be beneficial if the husband is nevertheless still firmly convinced that he is not its father.[27] It can be strongly argued that on balance it would be better for the child if it was firmly established who his father was rather than to leave this in doubt, even if leaving it in doubt secured for him the legal status of legitimacy. Also, it is to be remembered that the fact that a child is illegitimate at birth no longer means that he will remain illegitimate for all time. Since the Legitimacy Act 1959, the subsequent marriage of the natural parents legitimates a child even though one or both of them were married to another at the time of its birth.[28]In a great many cases the wife, after divorce proceedings, marries the corespondent, thereby legitimating any child she has had by him. This is likely to be far more beneficial to the child than to leave it in a position in which, while it is in law a child of the previous (dissolved) marriage, none of those concerned really believes that it is in fact such.
  20. We suggest that today it is more important for the court to arrive at a right decision than for a child to be declared legitimate at all costs. And at present the costs often include the injustice of making husbands maintain children who are probably not their own and the disrepute into which the law is brought as a result. We agree with the view of Professor Cross, expressed in a broadcast talk,[29] that the danger of injustice to husbands justifies the reduction of the standard of proof required to rebut the presumption of legitimacy. We see no good reason why the presumption should not be rebuttable on a balance of probabilities to accord with the standard of proof required in affiliation proceedings.[30] We do not, however, think that the presumption of legitimacy itself should be abolished; in our view the burden of proof should remain on a person seeking to bastardise a child born in wedlock. The burden would simply be less difficult to discharge if our recommendation were to be accepted.
  21. Our concern about the standard of proof required to rebut the presumption of legitimacy stems from our view that the criminal standard of proof can, if applied strictly, lead to the courts reaching unjust decisions. We have already quoted the remarks of Barnard J. in Watson v. Watson[31] which illustrate this danger, but an even more striking case is one from New Zealand. In Ah Chuck v. Needham[32] the child was a clear Mongoloid half-caste, though the husband and wife were both white. At the time of the child's conception the wife had been on intimate terms with a Chinese man (the co-respondent) but the husband was unable to prove non-access and the presumption of legitimacy was applied.[33]
  22. We have seen that the strictness of the presumption of legitimacy was due to society's reaction to illegitimacy. Similarly the matrimonial offence of adultery was regarded as hardly less serious than a criminal offence[34] and for this reason had to be proved in the courts with the same strictness. In 1948 this rule was re-affirmed by the Court of Appeal in Ginesi v. Ginesi[35] and shortly afterwards the House of Lords assumed this to be correct.[36] However, it is clear that if the standard of proof required to rebut the presumption of legitimacy is to be reduced then the standard of proof required to prove adultery must correspond. It would be absurd for a court to hold that evidence before it was sufficient to rebut the presumption of legitimacy but insufficient to prove adultery. Since, in England, adultery is not a criminal offence we see no reason why it should not be established on a balance of probabilities like any other non-criminal act. Fortunately the view upheld in Ginesi v. Ginesi mentioned above seems no longer to be the law.[37] Recently, in Blyth v. Blyth[38] it was suggested in the House of Lords that matrimonial cases should not be regarded as analogous to criminal proceedings and obiter dicta of Lord Denning and Lord Pearce rejected Ginesi v. Ginesi in strong terms.[39] On the basis of these dicta Rayden on Divorce accepts that the effect of Blyth v. Blyth is that "as far as the standard of proof is concerned, adultery, like any other ground for divorce, may be proved by a preponderance of probability."[40] Lord Denning M.R. has held in three cases since Blyth v. Blyth that the criminal standard of proof does not apply to adultery.[41]
  23. (II) The Present Powers of the Court
  24. The lightest exercise of physical violence to the person, even though beneficial (e.g. a surgical operation[42]), is an assault in law unless it has been consented to or unless there is some other lawful justification. In this technical sense the taking of a sample of blood for testing constitutes an assault in the absence of consent or lawful authority such as a court order.[43] Where a person lacks capacity to give a valid consent, any blood test on him, unless ordered by a court, will amount to an assault unless those who can consent on his behalf have done so. The two main categories of people who are not of full capacity are persons suffering from mental disorder and infants.
  25. There is no clear authority to show whether a voluntary mental patient who is able to understand the nature of a blood test can validly consent to it without the concurrence of some other person, although in Bolam v. Friern Hospital Management Committee[44] the patient's consent alone appears to have been accepted as sufficient for the administration of electro-convulsive therapy. Where the patient is not a voluntary patient, but is compulsorily detained, the position is obscure, owing to the lack of authority on the subject.[45] If the patient is capable of understanding the implications of a blood test it is not clear whether his consent alone is sufficient or whether it is necessary to obtain the consent also of the person having control of him. This might be the medical superintendent, the guardian or, in some cases, the nearest relative having power to discharge the patient. Where the patient who is subject to control does not appear to understand the nature of a blood test it may be that no one can consent on his behalf. The difficulty in this last case is that a blood test is not medical treatment and the court might hold that the present rules relating to consent for medical treatment do not apply to blood tests. It is even less likely that the power of the court, under ss.101 and 102 of the Mental Health Act 1959, to administer a patient's "property and affairs" would extend to consenting to blood tests.
  26. Since the Court of Appeal drew attention in Bickley v. Bickley^ to the importance, whenever the paternity of a child of the family is called in question, of considering whether steps should be taken under rule 56 of the Matrimonial Causes Rules 1957 (now rule 108 of the Matrimonial Causes Rules 1968) to have the child separately represented, the Official Solicitor has appeared far more frequently as guardian ad litem for infants. The Official Solicitor takes the view that a guardian's consent is required until the infant attains the age of 21. However, the Medical Defence Union[47] has suggested that the law does at present allow an infant who has reached the age of 16 to give a valid consent to surgical treatment and Mr. David Lanham in an article entitled "Blood Tests and the Law"[48] argues that the age of 16 should be that at which an infant can validly consent to a blood test, "since this appears to be the age which the medical profession take to be the age of consent", though he concedes that a doctor would be wise to obtain the guardian's consent as well where the infant is under 21.
  27. Although strictly outside its terms of reference the Committee on the Age of Majority, which reported in July 1967, considered the question of consent to medical treatment. In its Report [49] the Committee said:
  28. "The question of consent to medical treatment was not one which we had expected to consider, but so many witnesses begged us to clear up the law on this point that we felt bound to deal with it. It certainly seems that there is a good deal of confusion about what the legal situation really is, and that, even where separate witnesses felt they were in no doubt about the law, they had considerable misgivings about the consistency with which it was put into practice."[50]
  29. The Committee went on to recommend that persons aged 16 and over should be able to consent to medical treatment and that such consent should be as valid as the consent of a person of full age.[51] It gave, as its reasons for choosing the age of 16 as the right one for consent to medical treatment, the following:
  30. "(l)By National Health Service (General, Medical and Pharmaceutical Services) Regulations 1962, a person over 16 may choose his own doctor, and therefore presumably accept such treatment as his doctor advises. Similar regulations apply to dental and ophthalmic treatment.
    (2) By the Mental Health Act 1959, a person of 16 or over may make decisions for himself regarding admission as a voluntary patient to a mental hospital, and may apply to a Mental Health Review Tribunal.
    (3) 16 is the age of consent to sexual intercourse.
    (4) The practice of accepting consent to medical treatment at 16 has now been widely used for a considerable period, and has apparently proved satisfactory."[52]
  31. The courts have power to order a physical examination of the parties in certain cases. The most important of these (fairly closely analogous to the power to order a blood test) is the ancient and undoubted power of the court to order a medical inspection in nullity cases where impotence is alleged. In former times disobedience to the court's order to undergo an examination appears to have been regarded as contempt of court, and as late as 1842 in Harrison v. Sparrow[53]the writ de contumace capiendo issued to the sheriff with a view to punishing a disobedient husband with imprisonment. In the event he was not imprisoned and he was allowed to prosecute his appeal; and it seems unlikely that any party would be imprisoned today for disobedience to the court's order for an examination. On the other hand refusal to submit to a medical inspection may be treated by the court as some evidence of incapacity.[54]
  32. The power of the ecclesiastical courts to order medical inspection derived from the inherent jurisdiction of the court. The present position is governed by s.32 of the Supreme Court of Judicature (Consolidation) Act 1925 which prohibits reliance on the old practice of the ecclesiastical courts in matters of procedure and practice in so far as such matters are provided for in that Act or in rules of court. The court's power in the latter respect now rests entirely on rule 30 of the Matrimonial Causes Rules 1968 replacing rule 24 of the Matrimonial Causes Rules 1957.[55]
  33. Any power of the courts to order blood tests is not, of course, "ancient" since it was not until 1900 that Landsteiner announced his discovery of blood groups. The first use of blood group evidence in England appears to have been in a criminal case in 1929 ;[56] during the next decade it began to be used in civil cases involving paternity disputes.[57] It would seem clear, therefore, that the courts cannot derive power to order a blood test from the practice of the old ecclesiastical courts whose jurisdiction in these matters was brought to an end in 1857. In 1963 the Court of Appeal, in W. v. W. (No. 4),[58] decided, in a nullity suit, that the court had no power to order blood tests. However, in 1967 the Court of Appeal, in Re L.,[59] held that the court does have power, in custody proceedings, to order the child concerned to be tested. In Re L. the Court of Appeal upheld Ormrod J's decision in the lower court[60] that, because the Chancery Division could, in the exercise of its wardship jurisdiction,[61] order an infant to be blood tested, so a judge of the Divorce Division could order an infant to be tested where the court was exercising a parental jurisdiction (i.e. in proceedings such as wardship or custody); the effect of the Judicature Act 1873 was to give all judges of the High Court the powers exercisable in any Division.
  34. The Court of Appeal's earlier decision in W. v. W.[62] was distinguished on the ground that the ratio decidendi in that case was that a blood test would amount to an assault unless consented to or authorised by law and that no such authority could be found in any enactment or rule applicable to that case. The special powers of the court in relation to children were not considered in W. v. W. and that case was therefore not authority against the existence of the power of the court, in the exercise of its parental jurisdiction, to order a child to be blood tested. As the court did have this power a blood sample taken pursuant to an order of the court would be a lawful act and W. v. W. would not apply. Counsel for the Official Solicitor argued that a decision that the court had power to order blood tests in custody proceedings would produce anomalous results in divorce proceedings where the legitimacy of a child was questioned. Unless the court adjourned the divorce proceedings and dealt with the issue of custody at once, it might be constrained by the presumption of legitimacy to find the child legitimate in the course of the divorce hearing, although blood tests might prove it to be illegitimate in the course of the later hearing of a custody application.
  35. Shortly after the decision of the Court of Appeal in Re L., Wrangham J. held,[63] on an issue to determine the paternity of a four year old child, that, although he felt that a blood test would be in the best interests of the child, he had no power to order one as the issue did not involve the parental jurisdiction of the court. However, Lane J., in Hardcastle v. Hardcastle,[64] ordered a blood test although the issue was simply adultery. This divergence of judicial opinion was considered in the Court of Appeal in B.R.B. v. J.B.[65] Here again the question to be decided was whether the court could order a child to be blood tested where there was no question of the exercise of the court's custodial jurisdiction but where the issue was whether or not the wife had committed adultery. The paternity of the wife's child would determine whether she had committed adultery and for this purpose it was necessary that blood tests should be made. Lord Denning M.R. held that there was no reason for confining the court's power to order blood tests, as Willmer L.J. had in Re L.,[66] to the court's custodial jurisdiction. "The jurisdiction is unlimited in a judge of the High Court. He can order a blood test to be taken whenever it is in the best interests of the child."[67] Diplock and Sachs L.JJ., the other two judges in the case, also came to the same conclusion.
  36. As a result of the decisions of the Court of Appeal in Re L. and B.R.B. v. J.B. any judge of the High Court may order a child to be blood tested whenever it is in the best interests of the child for this to be done. This is an important development in the use of blood tests by our courts but, for the reasons which follow, we do not think that it goes far enough to obviate the need for legislation:
  37. (a) The House of Lords has yet to consider the issues raised in W. v. W. (No. 4), Re L. and B.R.B. v. J.B. and to this extent there remains an element of uncertainty in these decisions. This is particularly so regarding B.R.B. v. J.B., which decided that the power to order a child to be blood tested was not limited to the exercise of the court's custodial jurisdiction. In Re L. Willmer L.J. (with whom Arthian Davies L.J. expressly concurred) had said:[68]
    "I confine my judgment with regard to the court's power to order a test of a child's blood to cases arising, as this case does, within the court's custodial jurisdiction. I am not, as at present advised, prepared to hold that such a power exists in a paternity issue, and still less on a petition for divorce on the ground of adultery. So to hold would in my judgment be contrary to the decision of this court in W. v. W. (No. 4)."
    (b) The courts cannot order adults to be blood tested.[69]
    (c) The power of the court to order children to be blood tested is exercisable only by the High Court. The county court and the magistrates' court do not exercise parens patriae jurisdiction and therefore cannot order tests.[70] This is of particular importance now that undefended divorces are heard by county courts—though the need for blood tests is less likely to arise if the case is undefended.
    (d) As we have already mentioned, in paragraph 15 above, we think that it is important that the burden of proof required to rebut the presumption of legitimacy should no longer be the strict criminal standard of proof beyond all reasonable doubt.
    (e) If blood tests are to be used by the courts it is important that the necessary machinery should be set up to deal with administrative and procedural problems which will arise and this can only be done by legislation.
  38. The powers of the courts to use blood tests, in any civil proceedings, should be clearly spelt out and the development of this branch of the law should not be left entirely to the courts. We are anxious that the courts should not find themselves in the type of predicament which is illustrated by the remarks of the judges in two recent divorce suits. In one case evidence from blood tests was available to the court but in the other the mother refused to submit to a blood test and no evidence from blood tests could therefore be put before the court. In the first, Holmes v. Holmes,[71] Ormrod J., after finding that the wife's adultery had been proved on the evidence of blood tests made on the parties and the child whose paternity was in issue, said:
  39. ". . . had difficulties been put in the way of the child's blood being taken, it is manifest on the facts of this case that a grave injustice might have been done. It would have been virtually impossible upon the evidence, I think, for this man to establish that he was not, prima facie, the father of this child. He was living with the mother; it is true that he was not there at the crucial moment, but making all allowances for errors of calculation in periods of gestation and so on it would have been a very close run thing for him, with the presumption that the child was his child—because it was born in wedlock—heavily against him."
    "When, as I think in these days, it is possible to enable the Courts to do justice on a footing of fact and not to do injustice on a basis of presumption, I should myself greatly hope that no difficulties will ever be put in the way of a child's blood being supplied for blood grouping. I know it is a sad thing to bastardize a child, but there are graver wrongs; and I think that this is a matter which I am sure all those concerned will approach with greater caution because I think there is nothing more shocking than that injustice should be done on the basis of a legal presumption when justice can be done on the basis of fact . . ,"[72]
  40. Those remarks were echoed by His Honour Norman Richards, Q.C., sitting as a Special Commissioner in the second of these cases, Tarran v. Tarrant[73] He said:
  41. "I take the view that there is no obligation on the wife or the child, as the law is at present, to undergo a blood test and, as she is entitled to say no (although her reasons may not be such as commend themselves to the court), the court is not entitled to draw an inference adverse to her case on her refusal. But even more so, I take the view that the court in a matter of this kind, where the question of legitimacy is concerned, is not entitled to draw an adverse inference on that particular issue from the wife's refusal where the child is affected."
    "I share the views expressed by Mr. Justice Ormrod to the effect that procedurally, the Rules should be altered ... so that there was an obligation on the party to provide for a blood test to be taken in these matters so that a just result might be arrived at. This very often could only be arrived at with the assistance of blood tests because it is clear that the evidence available to the court must of necessity be extremely sparse and very often confined to that of the two spouses on an issue of this kind. Further, as I think, it follows from these observations that such legislation should include a right for the court to draw an inference on refusal of a wife—or a husband for that matter—to permit the appropriate tests to be made; but until that happens, in my judgment the court is not entitled to draw such inferences."[74]
    (a) The Presumption of Legitimacy
  42. For the reasons which we have already given, in paragraphs 11 to 17 above, we recommend that the presumption of legitimacy be made rebuttable by proof on a balance of probabilities. If this is done we think that, to avoid any doubts in the future, it should be made clear that the presumption of illegitimacy[76] is rebuttable in the same way.
  43. (b) Blood tests undertaken voluntarily
  44. If, in any proceedings, under the law as it stands now, the parties agree to submit themselves and the child to a blood test before there is any question of the appointment of a guardian ad litem the results of these blood tests are admissible in evidence. We think that no obstacle should be put in the way of the continuance in future of a practice which will tend to reduce the questions in dispute in litigation and cut out a number of contentious proceedings. It would be desirable for such voluntary testing to be carried out in the way which we recommend later in this Report.[77] There will clearly be cases where this procedure cannot be complied with, for example, where one of the parties is abroad and the tests on his blood are done in the country where he is living, but in general we consider that it is important that the recommended testing procedure should apply wherever this is possible.
  45. (c) Power to direct the use of blood tests
  46. In all civil cases in which the court has to determine the paternity of any child as a question of fact, it should have the power to direct that the parties to the action submit to blood tests and that the child concerned and its mother be blood tested, even though they are not parties to the action.
  47. The power to direct blood tests should be confined to persons (apart from the mother and the child concerned) who are parties to the action. It is essential, in our view, that anyone who is to be blood tested as a possible father should have the general protection of being made a party to the action.
  48. (d) Joint defendants
  49. In paragraph 29 of our Working Paper we said:
  50. "If the court is not given the power to order blood tests on persons other than the child concerned and the parties to the action, then we suggest that the present procedure in affiliation proceedings should be given careful consideration. We are concerned with the position of the woman who, for example, knows that the father of her child must be one of two men but does not know (and has no means of herself discovering) which of the two it is. We suggest that the difficulties facing a complainant in cases such as Sinclair v. Rankin and Robertson v. Hutchinson, which we have already mentioned,[78] could be largely overcome if a complainant could take affiliation proceedings against more than one defendant, relying on blood-group evidence to indicate (if possible) which of them is the child's father. We have already suggested the possible value of blood tests in such a case. Similarly a man against whom an affiliation claim is made who has reasonable cause to believe that there may have been another man or men should have the right to join the other or others. Our proposals would involve a fairly radical change in the present character of affiliation proceedings and we foresee that a number of difficult problems may have to be solved. These matters, however, are of some importance and we think that the attempt should be made."
  51. Much as we were attracted, initially, to the idea that a complainant should be able to take affiliation proceedings against joint defendants and that a defendant should be able to join other men as joint defendants, we do not now recommend these changes in the law. Both proposals would involve radical changes in the nature of affiliation proceedings which would present many practical difficulties. If a complainant were to take proceedings against joint defendants she would be admitting having had intercourse with more than one man and the court's task would be to determine from which act of intercourse conception resulted. If it proved impossible to serve process on one of the defendants or if one defendant failed to attend the hearing the court would be unable to determine the issue of paternity. Moreover, a non-exclusion result in respect of more than one defendant would almost always defeat the complainant's case. There is one other practical difficulty with joint defendant proceedings which would, on its own, persuade us not to recommend their introduction. This is the possibility of defeating the purpose of the proceedings by a tactical refusal to be blood tested on the part of two or more of the joint defendants concerned. We think that it would be so easy to defeat the purpose of joint defendant proceedings in this way that their introduction would not achieve anything in practice. Two hypothetical cases will illustrate what we have in mind: A complainant issues an affiliation summons against A and B, as joint defendants. The court directs A and B to submit to blood tests but both refuse to be tested. We recommend later in this Report [79] that no one should be physically compelled to give a sample of blood for testing but that the court should be able to draw whatever inferences it thinks right from the refusal of a person to comply with a direction for blood tests to be made. But in the example we are using what inference can the court draw from the refusal of both A and B ? The most that could be inferred is that A and B both think that they might be the father of the child concerned, but that will not help the court to decide which of them is in fact the father. Similarly, if a complainant issues an affiliation summons against A, who joins B as a second defendant, and both A and B refuse to be tested, no inferences as to the child's paternity could be drawn from their refusal. We therefore recommend that no change should be made affiliation proceedings in this respect; joint defendant proceedings should not be
  52. One particular suggestion which has been made to us is that if it can be proved that more than one man had sexual intercourse with the mother, within the possible period of conception, but it cannot be proved which of the men is the father of her child, then all the men concerned should be made to contribute towards the maintenance of the child. This system of duty of support without the establishment of paternity operated in Norway for many years but was abolished by legislation in 1956 for reasons discussed by Professor Arnholm in an article entitled "The New Norwegian Legislation relating to Parents and Children."[80] The following passage from this article gives in our view, compelling reasons why we should not introduce a similar provision into our law:—
  53. "The part of the Act (of 1915) which caused most criticism as time went by was that containing the rules providing for the establishment of a duty of support unconnected with paternity . . . The 1915 Act—much against the intention of the legislature—came to depress the social position of those children whose right of support was granted without the establishment of paternity. Such a decision involved an assumption of the sexual promiscuity of the mother during the period of conception and the scheme of support served to remind the child of this very fact during the whole of its adolescence. This means placing a severe psychological strain on the child. Experienced social workers affirm that children settle down more easily where no duty of support is imposed at all. The child can then find refuge in the thought that the mother has only had sexual relations with one man, who has deserted her and cannot be found. Against the scheme of imposing on several men the duty of supporting the same child particularly sharp criticism was forthcoming. From an economic point of view, of course, it might be advantageous to hold several persons jointly liable. But the advantage was dearly bought. It involved a particularly brutal reminder of the mother's lapse."

    For very much the same reasons, Denmark, in 1960, also abolished the "duty of support unconnected with paternity".

    (e) Refusal to be tested
  54. Until the court makes a direction for blood tests to be made the guardian ad litem, where one has been appointed, has the right to make submissions against the direction being made. Once a direction has been made the guardian ad litem has no right to prevent it being complied with though he can, of course, appeal against it. The position of the guardian ad litem was made clear by Lord Denning M.R. in Re L.[81] where he said:
  55. "[Counsel for] the Official Solicitor submitted that the Official Solicitor as guardian ad litem was the representative of the child; and it was his responsibility and his alone to decide whether the child should be subjected to a blood test. If he refused to consent to it, the Court could not, he said, overrule his decision. I think that this contention mistakes the position of a guardian ad litem. He has not the custody of the child in the conduct of the suit. His decisions do not rest in his unfettered discretion. He must on occasion seek the approval of the Court, as for instance if he makes a compromise on behalf of the infant. Likewise, if the Court thinks that the child should go to a particular school or have a surgical operation, the guardian cannot say 'Nay'. So also if the Court thinks that the child should have a blood test. The guardian should seek the ruling of the Court on the matter and abide by its decision."[82]
  56. A more important question is how the court is to treat the refusal of one of the individuals concerned to submit to blood tests. We do not think that it would be acceptable to public opinion in general or to the medical profession in particular to exert physical compulsion in order to obtain blood samples. We recommend, therefore, that no blood sample should be taken from a person under a direction of the court without that person's consent or, if he is incapable of consenting, without the consent of someone entitled to act on his behalf in accordance with paragraphs 48 and 49 below. Equally, we do not think that anything would be achieved by giving the court power to fine, or apply any similar sanction to, a party who refuses. We would recommend that refusal be considered by the court as evidence from which it can draw whatever inferences it thinks warranted in the particular case. For this reason we recommend that the court should be given the power to make a direction for the use of blood tests rather than an order requiring them to be made. We think that the court should not, in this context, make a formal order which would automatically attract penalty provisions such as those in s. 54 of the Magistrates' Courts Act 1952.
  57. In our Working Paper[83] we suggested that a person refusing to be tested ought to be able to show good cause, on religious or health grounds, why his or her refusal is justified. We still think that it should be open to a person refusing to be tested to satisfy the court that his or her refusal is justified but we do not think that it would be appropriate to provide specific grounds on which refusal can be justified. In the first place it would be difficult to anticipate all the reasons which would justify refusal (religious objection and health hardly seem to be exhaustive), and secondly the two grounds which we did suggest seem to be inappropriate. We have not been able to discover a religious body which objects to the removal of a sample of blood from one of its adherents. So far as health is concerned, we are advised by the medical profession that there are hardly any cases where a person's state of health would make it dangerous to have a blood sample taken, though there may be cases, such as with haemophiliacs, where precautions are necessary.
  58. It has been suggested to us that a provision similar to those in the Road Traffic Act 1962[84] and the Road Safety Act 1967[85] relating to the refusal of the accused to supply a sample of blood or of breath with a view to determining the alcohol content in his blood, should be applied to the refusal to be blood tested. It would be quite proper in our view to provide expressly that a person should be entitled to refuse to be tested with impunity if he satisfied the court that his refusal was, in the circumstances, justified; we think, however, that it would be unnecessary to do so if our recommendation is accepted that the court should be able to draw whatever inferences it thinks proper from a refusal. If the court regards the refusal as justified it will no doubt think that no inference should be drawn against the person refusing.
  59. The power of the court to draw inferences from a refusal to be tested may lead to problems in cases where it is inappropriate to draw any inferences on the issue of paternity. The following case, discussed in paragraph 37 of our Working Paper, illustrates the type of case which we have in mind:
  60. A husband petitions for divorce, alleging that his wife has committed adultery and maintaining that he is not the father of her child. He establishes a prima facie case on the issue of adultery and, at his request, the court orders blood tests. If the wife refuses to be tested she runs the risk that the court may infer from her refusal that she has committed adultery. On the issue of adultery therefore she will have nothing to lose by refusing, if she is herself convinced that her husband will be able to prove her adultery without having to rely on an exclusion result from blood tests. However an exclusion result will have the effect of bastardising her child and she may well not be prepared to run this risk, not only because she wants the legitimacy of her child upheld, but also because she wants to obtain maintenance for the child from the husband. To dissuade the wife from refusing to be tested for this tactical reason, ought the court to be able to infer from her refusal that he child is illegitimate ? Let us suppose that she herself does not know whether the husband or the co-respondent is the father of her child; both had intercourse with her during the possible period of conception and either could be the father.
  61. So far as adultery is concerned the wife knows, as a fact, whether or not she has committed adultery. It is, therefore, proper that the court should be able to infer from her refusal to be tested that she is trying to prevent it from discovering a fact of which she, herself, has knowledge. On the question of the
  62. "child's paternity however the position is different. Here she does not know which of the two men is the child's father; she is refusing to be tested, not in order to hide facts of which she has knowledge, but because she does not want to run the risk of her child being shown to be illegitimate. Her refusal has no bearing on whether or not the child is illegitimate and the court may well decide that it cannot draw any inference on the issue of paternity.
  63. The problem is not confined to cases where the child's mother refuses to be tested. A husband can also take advantage of the presumption of legitimacy and refuse to be tested if he is anxious to have the child's legitimacy established. Let us take as an example the case where a husband is divorcing his wife on the ground of her adultery. The wife admits her adultery and it is clear that both the husband and the co-respondent had intercourse during the period of conception of the child concerned. The husband and the wife are each applying for custody of the child. There is no way, apart from using blood test evidence, by which it can be established which of the two men is the child's father.[86] The wife, wanting custody and wanting to prove that the co-respondent, whom she is to marry, is the child's father, asks for blood tests. The husband, also wanting custody, refuses to be blood tested because he wishes the child to be declared his. The court cannot properly draw from his refusal the inference that the child is illegitimate; neither the husband nor anyone else knows who is the child's father. The presumption of legitimacy is therefore applied and the husband is held to be the child's father.
  64. How can parties be prevented from refusing to comply with a court order for blood tests, purely as a matter of tactics, so as to prevent the presumption of legitimacy being rebutted? We have already said that we do not think that it would be acceptable to force a person by physical compulsion to submit to an order for blood tests to be taken. We think that it would be equally unacceptable to treat a refusal as contempt of court; in any event there is a limit to the effectiveness of the sanctions which can be imposed for contempt, particularly in this context. If a wife is determined to obtain maintenance from her husband (say £3 per week for some 16 years) she might well decide that a £100 fine or a month or two in prison would be an acceptable price to pay. Moreover, it can rarely be in a young child's interests to imprison its mother. The answer is to find some means of preventing a tactical advantage being gained by refusing to be tested. One way in which this could be done would be to provide that:
  65. Where paternity is disputed and a party has refused to comply with a direction of the court for blood tests to be made, then, on an application by that party for custody of, or maintenance for, the child, the burden will be an the party applying (until the direction for blood tests is complied with) to prove the paternity of the child.

    In effect, the presumption of legitimacy would be suspended until the court's direction was complied with. If, for example, the wife refuses, she would have to establish that the child was her husband's; if she complied the presumption of legitimacy would operate.

  66. The difficulty with this solution is that it would amount to the introduction here of something very close to a "presumption of illegitimacy", which would hardly accord with our expressed aim of regarding the child's interests as paramount in a paternity dispute.
  67. In our view the most effective way of dealing with this problem is to provide that where a direction for blood tests is made and a party to the proceedings is entitled to rely on the presumption of legitimacy in claiming relief, then if that party refuses to be tested the court may either draw inferences against him (if appropriate) or may dismiss his application for relief. The court should also be given the power to adjourn the proceedings, before dismissing the application, in order that the person refusing to be tested should have time to reconsider his decision and obtain advice. The court should be given a discretion not to dismiss the application, so that it can proceed without blood test evidence if it thinks that the refusal to be tested is reasonable. Even this solution will have its. imperfections, and we readily recognise them: it does not cover the case where some third party (e.g. a Local Authority) is trying to obtain maintenance from the husband and the wife refuses; furthermore, in some cases the dismissal of an application may harm the child. Nevertheless we think that in practice a provision in this form will effectively deter parties from refusing to be tested for purely tactical reasons; they will cease to gain any advantage from the presumption of legitimacy so long as they refuse.
  68. (f) Refusal by a child or mentally disordered person
  69. The child's consent to the taking of a blood test directed by the court produces a separate problem. In practice there will be very few cases involving paternity where the child is more than a few years old. In the case of affiliation proceedings the general rule is that proceedings must be brought within a year of the child's birth and in most cases[87] maintenance orders cease when the child is 16. In most divorce cases and nullity suits, where paternity is in issue, the child concerned will be well under the age of 16 and in the great majority of cases will be only a few years old. Unless the child is a party to the proceedings, which will rarely be the case, it would be meaningless to provide that refusal is to be evidence against it. Hence, we suggest that the following principles should govern this problem:—
  70. (a) Any child of the age of 16 or over should be capable of giving a valid consent to having its blood tested, unless it would not be capable of giving a valid consent even if of full age. We consider that a mentally normal child of the age of 16 is quite capable of appreciating the implications of submitting to a court's direction for a blood test to be taken.[88]
    (b) If the child is under 16 its consent will not be effective. As already pointed out in paragraph 38, once the court has made a direction for blood tests to be carried out it is not for the guardian ad litem to give or withhold his consent to the order being carried out.[89] That, however, will not necessarily lead to the blood testing being carried out since, even though a guardian ad litem has been appointed, either one of the parties or some other person (for example, an ordinary guardian) will have actual care and control of the child. The consent of that person will in fact be required, since it would not be practicable to compel the administration of a blood test without his or her concurrence. If one of the parties has care and control of the child and refuses to allow the child's blood to be tested such refusal should be admissible in evidence against the party so refusing. In the case of petitions for legitimacy declarations the court should be entitled to draw whatever inferences it feels are justified where the person having the right to decide on behalf of the child refuses. In the case of these petitions the child will be a party to the action and may well stand to benefit substantially in a financial sense from the result of the proceedings.
    (c) If the child is 16 or over and refuses to submit to a blood test the court should be able to take into consideration any evidence touching on the refusal, such as persuasion by one of the parties concerned, and draw any inference which seems justified.
  71. It may happen that a person who is to be blood tested under a direction of the court is mentally disordered and, as a result, incapable of giving a valid consent. Yet it may be important, both in the interests of justice and in his own interests, that his blood should be tested. It would be possible to provide that the court should have the power to dispense with the need for consent in such case provided that there were no medical reasons why a blood sample should not be taken. This would, however, have one serious defect; it would not cover the case where the court was unaware of the mental disorder. The doctor who is to take the blood sample may be the first to discover that the person concerned is mentally incapable of consenting and we think it important that the doctor faced with such a situation should have clear guidance as to how safely to proceed. We recommend that it should be provided that, where a person is mentally incapable of understanding the nature and purpose of blood tests, and the person who has care and control of him consents and the medical practitioner in whose care he is confirms that no harm will be caused thereby, a blood sample may be taken. If the court is aware that one of the persons concerned is mentally disordered it will, before directing the use of blood tests, no doubt either make sure .that the person having care and control consents and that the medical practitioner does not object to his patient being tested, or will make the direction in terms which specifically require that the relevant consents are obtained. If the court made a direction for blood tests without being aware that one of the persons to be tested was mentally disordered, then the doctor taking the actual sample of blood could insist on the relevant consents being obtained if he were not sure that the person to be sampled was capable of giving a valid consent himself.
  72. (g) Non-exclusion results as evidence
  73. In paragraph 5 we have indicated what blood tests can and cannot prove. In Appendix B we discuss further the scientific nature of blood tests and their application to the determination of paternity. It seems to be beyond dispute that the accuracy of these tests is proven, that where they are properly conducted they are fully accepted by authoritative medical opinion and that they can give a conclusive exclusion result. They can prove that a man could not be the father of a given child. In the present state of medical knowledge blood tests cannot prove that a man must be the father of a given child; they can at most show, with varying degrees of probability, that he could be. In paragraph 35 of our Working Paper we said:
  74. "Where blood tests show a possibility that the putative father could be the father of the child concerned, we consider that the court should be able to take this possibility into account, together with all other evidence available. Clearly the court would have to be very careful as to what value it placed on such evidence. In some cases any one of, say, 50 per cent of the adult male population of the country, as well as the putative father, could genetically be the child's father; here no weight would be given to this evidence by the court. But there would be other cases where the presence of some uncommon genetic factor in both putative father and the child would make the chance of there being other possible fathers minute. Where such extremely uncommon blood factors are involved, the court should be able to rely on the blood test as weighty, though not conclusive, evidence of paternity."
  75. It can, however, be argued that because of the essential difference in probative value between an exclusion and a non-exclusion result, only exclusion results should be admissible in evidence. The arguments which can be advanced for making non-exclusion results inadmissible in evidence are:—
  76. (a) In the present state of medical knowledge non-exclusion results will very rarely give a strong indication of paternity,- only in a minute proportion of cases can they be treated as tantamount to proof of paternity.
    (b) To obtain non-exclusion results which give a strong indication of paternity generally involves extremely refined techniques and extensive testing, requiring rare and expensive testing reagents.
    (c) To be of any value in establishing paternity a non-exclusion result would have to express the statistical possibility of there being other men in the population who could be the father of the child concerned. There is a danger that courts, working on the principle of proof on a balance of probabilities, would regard a non-exclusion result which showed that, statistically, the chances of the putative father being the actual father were more than 50 per cent as sufficient to prove paternity. In fact only a non-exclusion result which showed that, say, 1 per cent or 2 per cent of the male population could be the child's father ought to be regarded as positive evidence of paternity. There would be, in other words, a danger of the courts regarding non-exclusion results as evidence of paternity when there was no scientific justification for their doing so.
    (d) The courts would undoubtedly require expert evidence to assist them in deciding what weight to give to a non-exclusion result and this would both add to the overall cost of blood tests and also place a heavy, and time consuming, burden on the few expert serologists capable of giving such evidence.
  77. We think that these arguments can all be met and that non-exclusion results should be admissible in evidence:—
  78. (a) While it is true that non-exclusion results will rarely give a strong indication of paternity, nevertheless there will be cases where such results are obtained and they will constitute valuable evidence which ought not to be withheld from a court which is trying to establish a child's paternity. Moreover, a non-exclusion result may be the final confirmation of paternity no matter what the strength of the indication of paternity which it gives. In Re i.,[90] for example, it was accepted by the parties and by the court that the father of the wife's child could only be the husband or the co-respondent. In such a case an exclusion result in respect of one of the men coupled with a non-exclusion result in respect of the other would establish the child's paternity without doubt. The result would, no doubt, be the same without the admission of the non-exclusion result but it would obviously be more satisfactory for all concerned to have this additional information before the court.
    (b) While it may be costly and difficult to obtain a meaningful non-exclusion result in some cases, that is beside the point if a non-exclusion result is in fact obtained. The question with which we are concerned is whether that result should then be admissible, not with how expensive it was to obtain.
    (c) The fear that the courts may not appreciate whether a non-exclusion result is of any probative value in establishing paternity is ungrounded. A properly worded certificate of results can make it quite clear to the court that a particular non-exclusion result should not be relied on as evidence of paternity. Indeed, it would be much safer to admit non-exclusion results together with a statistical analysis of their relevance to proving paternity. The risk otherwise would be that a court, knowing that blood tests had been made and no results put in evidence, would conclude that a non-exclusion result must have been obtained and that this must mean that paternity was proved.
    (d) Experience both in this country and abroad shows that non-exclusion results do not give rise to the extensive cross-examination of serologists which might be feared. This is another potential problem which can be avoided by detailed and clearly presented certificates giving the result of tests.
    (h) Procedure in Affiliation Proceedings
  79. We feel that where a man is accused of paternity in affiliation proceedings he should have the right to seek the help of blood tests to establish that he is not the father of the child concerned, whatever the strength of the evidence implicating him. It may be that the complainant is able to make out a strong case against the defendant which is nevertheless entirely false, and that the defendant is able to offer little in the way of evidence himself beyond a straight denial that he is the child's father. Because of the danger of a perjured claim in this type of case we recommend that a man accused of paternity should not be denied the right to obtain evidence from blood tests which may, in some cases, be the only source of convincing evidence available to him. A complainant may be able to establish that the defendant had intercourse with her, and this may be true, but intercourse is not necessarily indicative of paternity. Someone else may in fact be the father and the defendant may justifiably suspect this, yet be unable to offer any convincing evidence to the court.
  80. We propose the following procedure:—
  81. (a) A summons under the Affiliation Proceedings Act 1957 will indicate to the defendant that he will have the right to ask for blood tests during the course of the hearing.
    (b) In a form served with the summons the defendant will be invited to state whether or not he intends to apply for blood tests. There will, of course, have to be a brief explanation of the purpose of these tests to enable the defendant to make his decision on an informed basis. The form will also notify the defendant that, if the complainant agrees, the tests may be carried out before the hearing.
    (c) If the defendant states that he will ask for blood tests, the court will immediately notify the complainant of this and, if she agrees, will direct that tests be carried out before the hearing. If she does agree, the sample-taking and testing procedure will be put into effect at once, and the hearing will be fixed for a date which will give ample time for the results of the tests to be available. It should not, in any event, take very long for these results to be obtained.
    (d) If the complainant does not agree to tests being made before the hearing, the defendant will be entitled to ask, during the hearing, for these to be carried out. The court may well decide to hear the complainant before deciding whether to direct the use of blood tests and it may, if the complainant's evidence does not show a prima facie case or is not corroborated, dismiss her application without directing the use of tests.
    (e) If, after hearing the complainant's evidence, the court is satisfied that she has made out a prima facie case it will explain to the complainant that it has power to direct the taking of blood tests already requested by the defendant and invite her to explain why she is unwilling for blood tests to be performed. At this stage in the proceedings, the court will have the opportunity to explain in detail to the complainant exactly what is involved in having these tests made and what they can and cannot prove. It will also explain to the complainant that, if she refuses to comply with a court direction, it will be entitled to draw whatever inferences it thinks right from her refusal, and may dismiss the case. There will thus be an opportunity for the woman to be persuaded by the court that her objections to blood tests are unfounded or irrational, and an opportunity for the woman to convince the court that her objections are well-founded. Then, in the light of the fullest possible information, the court will decide whether or not to make a direction.
    (f) If the court accepts that the woman's objections to blood tests are reasonable, it will proceed to a decision on the evidence that is available without blood tests. If, on the other hand, the court decides that her objections are without substance and should be overruled, it will make the direction and adjourn for a period, which would give time for her to change her mind and for the tests to be carried out. If, after the adjournment, the complainant has still refused to be tested the court will draw the appropriate inferences from her refusal and may well dismiss her application.
  82. We do not recommend that an exclusion result should be binding on the court in the sense that the court be expressly prevented from finding paternity proved in the face of an exclusion result. We think that an exclusion result should be treated merely as evidence to be considered by the court, along with all the other evidence which is available. However, we would not expect a court to find paternity proved in spite of an exclusion result, unless there was extremely strong evidence casting doubt on the validity of the exclusion result. In most cases where the validity of blood test results was in question a fresh set of tests would resolve any doubts, but it is conceivable that there might be reasons, in rare cases, justifying the court ignoring an exclusion result.[91] It follows that we do not recommend that a complainant should be prevented from continuing with her case despite an exclusion result in respect of the defendant.
  83. The costs of blood tests should be borne by the defendant initially and at the end of the proceedings the court would make an order as to who ultimately is to bear the cost. It is, of course, extremely important that the centres carrying out the tests should be guaranteed payment and for this reason we consider that the defendant, at whose request tests will be made, should be responsible initially for their payment.[92] The costs should be regarded as legitimate disbursements to be borne by the Legal Aid Fund in legally aided cases and it should not be necessary to obtain authority from the Area Committee before asking for tests.
  84. (i) Proceedings other than affiliation
  85. In divorce or nullity proceedings, and in any other civil proceedings in which paternity is in issue, the court should have a discretion whether or not to direct the use of blood tests at the request of any party. In divorce or nullity proceedings blood tests would generally be requested by a petitioner who was attempting to bastardize his wife's child to enable him to prove thereby his wife's adultery, to avoid having to maintain the child, or to obtain an annulment of his marriage on the ground that at the time of its celebration the wife was pregnant by another man. Blood group evidence could be used to support, and perhaps prove conclusively, the husband's case. If the petitioner were given the absolute right to require the court to direct the use of blood tests there would be a danger that petitions would be based on extremely slender evidence in the hope that blood tests would turn out to support the petitioner's case. For example, a husband might be encouraged to institute divorce proceedings on the ground of his wife's adultery, with no evidence to support the allegations in the petition, but in the hope that blood tests would confirm them. To deter speculative litigation of this nature we recommend that the court should be given a discretion whether or not to direct the use of blood tests.
  86. (j) Testing Procedure
  87. It is important that blood testing in connection with paternity disputes should be undertaken at properly equipped centres and that properly qualified serologists should be responsible for them. It is also important that uniformity in technique and testing reagents is achieved and that procedures are used which reduce the possibility of error to the absolute minimum. We recommend that special testing centres should be set up and that blood tests made under a direction of the court should be carried out only at these centres and in accordance with specified procedures. It will be necessary for regulations to be drawn up by the appropriate authorities (we recommend that these should be the Ministry of Health and the Home Office) specifying:—
  88. (a) the centres at which tests can be carried out,
    (b) the selection of experts to be responsible for the carrying out of the tests,
    (c) the nature of the tests to be made,
    (d) the procedure to be followed in making the tests, and
    (e) the form in which the result of the tests should be communicated to the parties.

    There will, of course, clearly have to be other detailed administrative rules and those we have mentioned above are by no means exhaustive.

  89. We have already recommended[93] that the results of blood tests undertaken voluntarily should continue to be admissible in evidence. It would be desirable for these to be carried out at the same testing centres and in accordance with the same procedures to be used for blood tests made under a direction of the court. We recommend therefore that the testing centres should be available to the public, on a fee paying basis, for voluntary blood testing in connection with paternity disputes. We do not think that it would be desirable to provide that blood test evidence should only be admissible if obtained from tests carried out at the testing centres, but we would hope that anyone wanting to obtain blood group evidence voluntarily would use the recommended procedure whenever possible. We anticipate that they would do so, in practice, because the validity of the results obtained from a testing centre would undoubtedly be less open to questioning.
  90. We suggest that the results of blood tests should be communicated to the parties and admissible in evidence in the form of a standard certificate, a copy of which should be sent to each of the parties concerned. The certificate should specify, together with reasons, whether the result of the tests is to exclude the putative father or to show that he could be the father. Where the certificate shows a non-exclusion result it should also state the statistical possibility of there being other men in the population who could equally well be the father and the serologist's opinion as to the value, if any, of the result in determining whether the person tested is the father of the child concerned. We do not suggest that the certificate should be treated as conclusive in the sense that it should prevent the expert conducting the tests from being called to give evidence before the court, though we would expect that detailed certificates would be accepted unchallenged in the vast majority of cases.
  91. Some practical difficulties may arise in connection with the interpretation of certificates, particularly in affiliation proceedings. It may be difficult for the relatively few expert serologists to be available if cross-examination of blood test evidence becomes a common feature of these proceedings, bearing in mind the number of magistrates' courts and their geographical distribution throughout the country. It may also be difficult in some cases for the magistrate to interpret the findings revealed by the certificate without further expert evidence which may not be called by the parties.
  92. We suggested in our Working Paper[94] that it might be useful to set up a Medico-Legal Council, comprising experts in the appropriate scientific fields, to give written opinions to the court in difficult cases. This is the procedure used in the Scandinavian countries. We do not recommend the setting up of a similar body in this country, primarily because the adversary nature of our civil proceedings requires that where evidence is in dispute the parties have the right to cross-examine the witnesses concerned. To provide that a written opinion from a Medico-Legal Council should determine the dispute would not only be a fundamental departure from the right of cross-examination, but would tend to substitute the decision of the Council for the decision of a court of law. Furthermore, we do not think that there would be a practical need for a Medico-Legal Council if certificates are presented clearly and with a printed explanatory guide. We do not expect that the interpretation of results will give rise to difficulties which could not be solved by the serologist concerned being called to give evidence in person. However, under the rule of evidence which prevents a party from cross-examining his own witness the party calling the serologist would be precluded from cross-examining him. Provision will clearly have to be made to enable the party calling the serologist to cross-examine him.
  93. It has been suggested to us that the risk of impersonation would detract from the evidential value of blood tests. One way of avoiding this danger would be for blood samples to be taken by a medical practitioner nominated by the court, all the parties attending at the same time and place so as to identify each other. However, we do not think that this would be practicable, particularly in the case of affiliation proceedings, for the parties will often be living far apart and it would be unreasonable to expect someone to travel perhaps hundreds of miles simply to have a blood sample taken. It would also be difficult, in many cases, to ensure that the parties did both attend for sampling at the same time. Furthermore, it might be that relations between the parties were so strained that it would be highly distasteful, and detrimental to the child, to insist that the parties and child attend together to be tested.
  94. Under the Matrimonial Causes Rules 1968 the following is the procedure employed in medical examinations in nullity cases:—
  95. "Every party presenting himself for examination shall sign, in the presence of the inspector or inspectors (i.e. the doctor), a statement that he is the person referred to as the petitioner or respondent, as the case may be, in the order for the examination and at the conclusion of the examination the inspector or inspectors shall certify on the statement that it was signed in his or their presence by the person who has been examined."[95]

    We think that for the purpose of identification in the case of blood tests it will be necessary for the Rules to provide for a rather more careful check on identity, particularly so far as the child is concerned. We would prefer an adaptation of the procedure previously used with nullity cases, under which the identity of the party to be examined was established by his solicitor accompanying him. If the party attended unaccompanied by his solicitor he had to produce a photograph certified by solicitors to be a true likeness and this was attached to the medical report.[96] In adapting this system to the administration of blood tests for the purposes of determining paternity it might well be necessary to provide additional safeguards to prevent substitution where very young children are concerned, e.g. by supplying a footprint as well as a dated photograph. We also recommend that it should be made an offence to impersonate a person whose blood is to be tested under a direction of the court or to proffer a child for testing knowing that it is not the child in respect of whom the direction was given.

  96. It is extremely important, to avoid error, that samples of blood should be correctly labelled and properly packed when being sent to the testing centre and that the samples should be taken using proper sterile equipment. For this reason we recommend that a panel of medical practitioners should be appointed in each area, and that blood samples given under a direction of the court should be taken only by members of a panel. We think that it would often be undesirable or embarrassing for a party's own doctor to have to take a sample and there are administrative reasons why a panel would be preferable. The procedure which we envisage is that on the court ordering blood tests the appropriate court official (i.e. the clerk to the justices in the case of affiliation proceedings) would instruct the parties to make an appointment with a nominated panel member; at the same time he would ask the testing centre to send the necessary equipment to the doctor concerned. This would ensure uniform and accurate labelling and the use of the proper sterile equipment. We think that a convenient service for the public could be maintained without the number of panels and of doctors on them being very large.
  97. We understand that in rare cases the accuracy of some of the blood group tests currently used can be affected if the person whose blood is being tested has recently suffered from certain illnesses or received a blood transfusion, though this should be detected in the course of conducting the tests. We therefore recommend that a person whose blood is to be tested under a court order should be required to make a statutory declaration specifying whether or not he or she has suffered from any illness during the preceding twelve months or has received a blood transfusion within the preceding four months.[97] No doubt the form and substance of this declaration will be settled in consultation with the appropriate medical opinion. The declaration should be submitted to the expert carrying out the tests, who should be required to comment on the facts disclosed and their effect, if any, on the tests made by him.
  98. We have already dealt with the question of costs in connection with affiliation proceedings.[98] So far as other proceedings are concerned we recommend that, as with affiliation proceedings, the person at whose request blood tests are ordered should be responsible, initially, for their payment. Again we would recommend that the court should have a discretion to decide who should ultimately bear the cost and the costs should be regarded as legitimate disbursements to be borne by the Fund in legally aided cases.
  99. It is difficult to give an accurate estimate of the number of cases, other than affiliation proceedings which we have already discussed,[99] in which blood tests would be ordered. We would not expect there to be a great number of tests made in connection with divorce proceedings as paternity disputes in these cases are relatively uncommon. As for nullity suits and legitimacy declarations, there were only 32 petitions in 1967 for nullity on the grounds of the wife's pregnancy by another man at the date of the marriage and petitions for legitimacy declarations number approximately the same. We are advised that a complete set of tests on three people (i.e. mother, child and alleged father) would cost between £20 and £25.
  101. (a) The presumption of legitimacy should be made rebuttable by proof on a balance of probabilities and it should be made clear that the same burden of proof applies to the presumption of illegitimacy. (Paras. 11 to 17 and para. 31. Clause 6 in Appendix A.)
  102. (b) The results of blood tests undertaken voluntarily should continue to be admissible in evidence. (Paras. 32 and 59. There is nothing in the draft clauses in Appendix A to prevent this evidence being admissible.)
    (c) In all civil proceedings in which the court has to determine the paternity of any child it should have the power to direct that the parties to the action, the child concerned and its mother submit to blood tests. (Paras. 33 and 34. Clause 1(1).)
    (d) Any party to the proceedings should be entitled to apply for a direction. (Paras. 53 and 57. Clause 1(1).)
    (e) No sample of blood should be taken from a person under a direction of the court unless that person consents to its being taken or, if he is incapable of consenting, unless consent is given in accordance with (h)-(j) below. (Para. 39. Clause 2(1).)
    (f) Where a person refuses to comply with the court's direction the court should be entitled to draw whatever inferences it thinks appropriate from the refusal. (Paras. 39 to 41. Clause 4(1).)
    (g) Where a person is applying for relief and is relying on the presumption of legitimacy, if he refuses to comply with the court's direction to submit to a blood test the court should have power to dismiss the application notwithstanding the presumption. (Paras. 42 to 47. Clause 4(2).)
    (h) A child aged 16 or over should be capable of giving a valid consent to giving a sample of blood unless, if of full age, he would not have the capacity to consent. (Para. 48. Clause 2(2).)
    (i) Where a child is under the age of 16 the consent of the person having care and control of him should be required. (Para. 48. Clause 2(3).)
    (j) If a person is mentally incapable of giving a valid consent it should be in order to take a blood sample from him if the person in whose care and control he is consents and the medical practitioner under whose care he is certifies that giving a sample will not be prejudicial to his proper care or treatment. (Para. 49. Clause 2(4).)
    (k) Both exclusion and non-exclusion results should be admissible in evidence. (Paras. 50 to 52. Clause l(3)(a) and (b).)
    (1) The party applying for blood tests should, initially, bear their cost but the court should have a discretion to decide who should ultimately be responsible for paying for the tests. The costs should be regarded as a legitimate disbursement to be borne by the Fund in legally aided cases. (Paras. 56 and 67. Clause 1(2).)
    (m) Regulations should be made governing the procedures for carrying out blood tests directed by the court. (Para. 58. Clause 3.)
    (n) The results of blood tests should be capable of proof by a certificate from the serologist responsible for the tests. (Para. 60. Clause 1(5).)
    (o) Provision should be made so that any party can call the serologist concerned to give evidence in person and can cross-examine him. (Para. 62. Clause 1.(4).)
    (p) Any person who impersonates another (from whom the taking of a blood sample has been directed by the court) or proffers a child in place of the child named in a direction should be guilty of an offence. (Para. 64. Clause 5.)
    (Signed) Leslie Scarman, Chairman. L. C. B. Gower. Neil Lawson. Norman S. Marsh. Andrew Martin.
    J. M. Cartwright Sharp, Secretary. 6th September 1968.

    Image 1

    Image 2

    Image 3

    APPENDIX B[100]
  103. The existence of blood groups, first demonstrated at the beginning of this century by Landsteiner, explained the hitherto unintelligible disasters (such as death or severe illness) which occurred frequently when blood transfusions were given to patients. Landsteiner found that when blood serum from one individual was added to samples of red blood cells from other individuals, in some cases, but not others, the red blood cells formed dense clusters—a phenomenon known as agglutination. He deduced from this that the red blood cells of some individuals contained different chemical substances from the red blood cells of others and that agglutination occurred only when the cells contained a chemical which was "incompatible" with the particular serum being used in the experiment. He found that he could classify all blood into four specific groups, termed O, A, B and AB, and that red blood cells from one group were either compatible or incompatible with the serum from the other groups according to a predictable pattern. Since Landsteiner's original discovery several other systems of blood groups have been found, including the MN and Rhesus systems. The substances which differentiate these groups cannot, as yet, be identified in terms of their chemical constitution but their presence or absence can be shown by the technique of agglutination which we have mentioned.
  104. Subsequently other types of tests such as the Hp and Gc tests have been evolved. With these the technique is entirely different, for complex proteins in the blood are separated out and identified by a process called electrophoresis. This process depends upon the fact that an electric field can cause the chemicals concerned to move through a medium such as starch gel and that they move at different rates, dependent on their molecular size and charge. Thus Hp.l takes up a characteristic position in the gel some distance from Hp.2 and the two substances can be separated from each other.
  105. The value of our knowledge of blood groups for the determination of paternity lies in the fact that the different factors present in each group are transmitted from one generation to another by the recognised principles of heredity. The mode of inheritance of blood groups has been established with a high degree of certainty by an enormous mass of research in many countries, involving many thousands of families, and the results of these experiments are completely in accord with the accepted rules of genetics. Without embarking on a detailed discussion of the mechanism of heredity a brief description can be given of how this mechanism applies to the inheritance of blood groups. In the nucleus of every normal human body cell there are 46 visually identifiable bodies known as chromosomes, arranged in 23 pairs. Apart from the chromosomes which determine sex, each chromosome of the pair is the same shape as the other. These chromosomes each carry a number of smaller bodies called genes and, put very simply, the transmission of every inherited characteristic from one generation to another depends upon the transmission of the corresponding gene or groups of genes. The human germ cells (i.e. ova and spermatozoa as opposed to normal body cells) contain only 23 chromosomes, only one of each pair of chromosomes from the normal 46-chromosome nucleus being used in the formation of the germ cell nucleus. Let us take, by way of illustration, a father who has the O factor in each of the relevant pair of chromosomes and a mother who has the A factor in one chromosome of the relevant pair and the O factor in the other chromosome of that pair. When the paired chromosomes divide in the formation of germ cells the father will produce germ cells which can only contain the O factor. The mother can, however, produce germ cells with either the A factor or the O factor, depending on which chromosome of the pair the germ cells take. The diagram below shows the possible combination of factors which the child of this mother and father can have, depending on which germ cell from the father fertilizes which germ cell from the mother. (It should be borne in mind that when a germ cell is fertilized by another germ cell the 23 chromosomes in each germ cell pair to give an embryo with the normal 46 chromosomes.)
  106. Image 4

    It can be seen that a child of these two parents cannot possess the B factor. If the mother has a child which possesses the B factor then its father must be a man whose chromosomes contain the B factor and cannot be the man in our illustration. In Part II we set out tables (dealing only with the ABO and MN systems) showing possible and impossible combinations of factors in children of parents whose ABO or MN groups are known.

  107. There is, theoretically, a possibility that in dividing to form germ cells the chromosomes may undergo a change in chemical composition so that, for example, a chromosome containing the O factor could change to possess the B factor instead. Clearly if this change, termed a mutation, were to occur it would invalidate the reasoning behind the diagram in the preceding paragraph. However, mutations in nature are known to be extremely rare and so far as blood factors are concerned only one possible example has been demonstrated in all the millions of cases investigated. Even that one is not regarded by the leading authorities in this country as more than fairly convincing.[101]
  108. A second fact which bears on the reliability of blood tests is that in any laboratory test there is the possibility of human observer-error. However, a vast experience of the techniques of blood grouping has been acquired in connection with the blood transfusion services in many countries. While observer-error is always a possibility, this can be virtually eliminated by repeating the tests on several samples of blood and the risk of observer-error is probably less than that involved in the identification of finger-prints.
  109. It is particularly important that the serologists conducting blood tests for the purposes of determining paternity are specially trained. Routine training in pathology and/or haematology and clinical pathology does not necessarily qualify one to do this type of blood testing. It cannot be overstressed that only properly qualified and experienced serologists should be appointed to carry out this work. In America, for example, serious errors have been found to occur because the person carrying out the tests was not sufficiently experienced.[102] It is also important that the standard of the materials used in testing should be carefully controlled. These considerations make it essential that testing should be carried out only at specified centres which employ expert serologists and use standard materials.
  110. A further source of possible error must be mentioned. Where, for example, the alleged father is of group MM (his blood reacting only for the M and not for the N factor) and the mother is MN, it is assumed that the child must be of group M or MN and cannot be N. This assumption depends, of course, upon a further assumption that there does not exist a third allelic gene, for the product of which gene no specific reagent or other means of recognition has yet been discovered. This assumption may, on occasions, be proved false. The discovery of a third allelic gene, now known as Mg, which gives rise to an antigen Mg not reacting either with anti-M or anti-N serum is a case in point.[103] Since errors of this type depend upon the existence of rare undiscovered factors, the errors themselves are bound to be rare. Furthermore they will usually tend to occur where exclusion of paternity is based on testing only the child and putative father, and not the mother.
  111. We have seen how blood groups can be determined and how the transmission of factors from one generation to another works in principle. Additional valuable evidence, so far as paternity findings are concerned, is provided by a statistical analysis of the distribution of factors in the population of any country. In Great Britain the distribution of the ABO groups is approximately:—
  112. O — 46 per cent
    A — 42 per cent
    B — 9 per cent
    AB — 3 per cent

    Statistical calculations show that using these groups alone the chances of being able positively to exclude a given man average about 17 per cent although if the child is group B or AB a greater proportion of men would be excluded as so few Englishmen have B to give.

  113. Since Landsteiner's original discovery and particularly since 1940, a considerable number of other blood groups have been discovered, i.e. a considerable number of other chemical substances have been shown to exist on the red cells. These substances are inherited independently of one another and so are described as different blood group systems. The relevant ones for determining parentage are set out below,[104]together with the cumulative chances of excluding a given person by determining the group of the child and of the mother and putative father, if all the available tests are employed.
  114.   Exclusion by each system (per cent) Cumulative Exclusion {per cent)
    1. ABO ... 17-6 17-6
    2. MNS ... 23-9 37-2
    3. Rh. (D, C, c, E) ... 25-2 530
    4. Kell(K) ... ... 3-7 54-8
    5. Lutheran (Lua) ... 3-3 56-3
    6. Duffy (Fya) ... 4-7 58-4
    7. Kidd(Jka) ... ... 20 59-6
    These tests alone offer, on average, a 60 per cent chance of exclusion. It must be remembered that this table applies only to inhabitants of Great Britain though its application to other Western Europeans produces broadly similar results.
  115. In individual cases the prospects of exclusion may be considerably higher than the figures in the table, if either the child or the putative father is found to have an uncommon blood group or a combination of uncommon groups. In extreme cases the chance of two unrelated men having the same combination of uncommon groups may be as low as 1-6 in one hundred million. In other cases the chance may be of the order of 1 in ten thousand or 1 in fifty thousand. In such cases proof that both child and putative father have the same rare or very rare combination is valuable positive evidence that the putative father is in fact the father.
  116. The blood groups mentioned in the table above are all based on chemicals found on the red cells of the blood. There are, in addition, other chemical substances which can be identified in the blood serum, i.e. the liquid component of the blood. These substances have also been shown to be transmissible from one generation to another in accordance with the rules of heredity and can therefore assist us in the determination of parentage. As we have already briefly stated, the techniques involved are quite different from those used in the identification of blood groups, but they are equally reliable in skilled hands. Two such substances are now being used in paternity cases and have been approved by the medico-legal authorities of Denmark and other Scandinavian countries. These are the haptoglobin[105] groups and Gc groups. It is possible to classify all samples of sera into three haptoglobin groups and three Gc groups described as Hp.1-1, Hp.1-2 and Hp.2-2 and Gc.1-1, Gc.1-2 and Gc.2-2. Approximately 15 per cent of the population of Western Europe are Hp. 1-1,47 per cent are Hp.1-2 and 36 per cent are Hp.2-2. The haptoglobins alone exclude 18 per cent of men erroneously alleged to be the father of the child and the Gc groups exclude 15 per cent. Since the Hp and Gc groups are inherited independently of one another and of the groups mentioned before, the combined exclusion rate if those tests are used also is raised from 59-6 per cent (see table above) to approximately 72 per cent. The haptoglobin and Gc tests have to be used carefully, for haptoglobins are not definitely developed in a child under three months old and ill health may sometimes make it difficult to identify these substances.
  117. Another system of blood grouping is by identification of inherited variants of the Gamma-globulin, discovered by Grubb in 1956. He showed that the blood sera of normal persons could be divided into two groups, Gm (a+) and Gm (a—) according to whether or not their serum prevented agglutination of anti-D coated rhesus positive cells by an antibody present in the serum of a proportion of rheumatoid arthritis sufferers and occasional normal individuals, ability to inhibit agglutination being inherited as a Mendelian dominant character. Other Gm groups have since been discovered. Gm(a) and Gm(b) have been used in evidence in paternity cases in Norway since 1962 and a third factor, Gm(x) has been employed in some countries. Most of the Gamma-globulin present in the newborn child is of maternal origin and it is not until the child is some months old that its Gm groups can be determined.
  118. We understand that phosphoglucomutase grouping is likely to be employed in paternity testing before very long. Here the lead is in this country.[106] This is an inherited system of blood-tissue enzymes which is already being used in anthropological studies and forensic identification tests. Grouping is by starch gel electrophoresis (like the haptoglobin grouping) followed by a special enzyme staining technique.
  119. Another group of substances present in blood serum, the lipo proteins, are at present under intensive study and it is possible that in the near future these will be valuable in the determination of parentage.
  120. Part II
    Parents' Blood Groups Their Children's Blood Groups
      Pos, dole Impossible
    O-O O   A, B, AB
    O-A O, A   B, AB
    O-B O, B   A, AB
    O-AB A, B   O, AB
    A-A A, O   B, AB
    B-B B, O   A, AB
    A-B O.A, B, AB None
    A-AB A,B, AB O
    B-AB B,A, AB O
    AB-AB A,B, AB o
    M-M M N, MN
    N-N N M, MN
    M-N MN M, N
    M-MN M, MN N
    N-MN N, MN M
    MN-MN M, N, MN None
    Man Chile
    O AB
    AB O
    M N
    N M
    Association of Law Teachers.
    British Institute of International and Comparative Law.
    British Legal Association.
    British Medical Association.
    British Society for Haematology.
    Catholic Marriage Advisory Council.
    College of Pathologists.
    Council of Married Women.
    Family Law Association.
    General Council of the Bar.
    Holborn Law Society.
    Institute of Forensic Medicine, Copenhagen.
    International Society of Haematology.
    John Hilton Bureau.
    Justices' Clerks' Society.
    Law Society.
    Lister Institute.
    Magistrates' Association.
    Married Women's Association.
    Medical Defence Union.
    Medical Research Council.
    Medico-Legal Society.
    Ministry of Justice, Denmark.
    Ministry of Justice, Norway.
    Ministry of Justice, Sweden.
    National Blood Transfusion Service.
    National Council for Civil Liberties.
    National Council for the Unmarried Mother and Her Child.
    National Council of Women of Great Britain.
    Newman Legal Studies Group.
    Pathological Society of Great Britain and Ireland.
    Scottish Law Commission.
    Society of Conservative Lawyers.
    Society of Labour Lawyers.
    Society of Public Teachers of Law.
    Printed in England for Her Majesty's Stationery Office by McCorquodale & Co. Ltd., London

Note 1    Published Working Paper No. 12, "Proof of Paternity in Civil Proceedings."    [Back]

Note 2   A list of the organisations whom we consulted is set out in Appendix C.    [Back]

Note 3    Law Reform Committee, Sixteenth Report (Privilege in Civil Proceedings), December 1967. Cmnd. 3472, paras. 10(c), 44 and 45.    [Back]

Note 4    Since we ourselves lack any expert knowledge of medical matters we have had to rely on others and are greatly indebted to Mr. Justice Ormrod (a qualified doctor), Dr. K. Henning-sen, Head of the Serological Department, University Institute of Forensic Medicine, Copenhagen, Dr. A. Grant, Lecturer in Forensic Serology, Guy's Hospital Medical School, Dr. A. S. Wiener of the Department of Forensic Medicine, New York University Medical School and Dr. A. R. Kittermaster, Consultant Pathologist, Kent and Sussex Hospital, for their help in compiling the scientific material in this Report. Any errors which we have fallen into in the course of consulting medical text books on the subject are attributable entirely to us.[Back]

Note 5    In Re L [1968] P. 119 the court accepted that in the circumstances, because it was known that the child's father must be one of two men whose blood groups were known, the chance of obtaining an exclusion result in respect of one of them, when the child's blood was tested, was 90 per cent. See per Lord Denning M.R. at 154.[Back]

Note 6    1921 S.C. 933 and see Robertson v. Hutchinson 1935 S.C. 708. In this type of case blood tests assist a woman who does not know which of two men is the father of her child.[Back]

Note 7    See e.g. B. v. Attorney General (N.E.B. and others intervening) [1967] 1 W.L.R. 776 where the petitioner prayed for a declaration that he was the legitimate son of N.E.B., as his legitimacy would have entitled him to share in a trust fund. Blood tests showed that the petitioner could not be the child of N.E.B. and the petition was accordingly dismissed, the petitioner offering no evidence.[Back]

Note 8    Cmnd. 3051 (1966).[Back]

Note 9    Brought under s. 1 of the Affiliation Proceedings Act 1957.[Back]

Note 10    See the Registrar General's Statistical Review of England and Wales for 1966. In 1954 the proportion of illegitimate births to the total number of births was 4-7 per cent; in 1964 it was 7-2 per cent and in 1966, 7-9 per cent.[Back]

Note 11    A woman who was a single woman at the date of the birth to her of a child or is single at the date of her application (and for this purpose a married woman living apart from her husband or a widow will, in certain circumstances, be deemed a single woman) may apply to a justice of the petty sessions area in which she resides for a summons against the alleged father.[Back]

Note 12    Affiliation proceedings constitute one of the few types of civil action where the evidence of a witness cannot be accepted without corroboration. The Scottish Law Commission, in its report of 16th February 1967 entitled "Proposal for Reform of the Law of Evidence relating to Corroboration", recommended that corroboration should be retained as a requirement in affiliation proceedings, for "by the nature of the case, caution has to be exercised in accepting the evidence of a woman raising an action of affiliation". The Law Reform (Miscellaneous Provisions) (Scotland) Bill now before Parliament, follows the Scottish Law Commission's recommendation on this point.[Back]

Note 13    Affiliation Proceedings Act 1957, s.4(2). [Back]

Note 14    Moore v. Hewitt [19471 K.B. 831.[Back]

Note 15    In affiliation proceedings the presumption will apply only where the child was born to or conceived by the complainant while she was married (see n.ll supra).[Back]

Note 16    (1811) 1 Sim. & St. 153, per Sir James Mansfield CJ.[Back]

Note 17   Head v. Head (1823) 1 Sim. & St. 150, per Leach V.-C. at 152.[Back]

Note 18    [1924] AC 687. The rule was initially propounded in Goodright d. Stevens v. Moss (1777) 2 Cowp. 591.[Back]

Note 19    Gordon v. Gordon [1903] P. 141.[Back]

Note 20    [1954] P. 48.[Back]

Note 21    At page 55.[Back]

Note 22    Re.L. [1968] P. 119.[Back]

Note 23    See para. 16 infra.[Back]

Note 24    Per Wfflmer L. J. at 171 and per Arthian Davies L.J. at 172.[Back]

Note 25    See para. 7 supra.[Back]

Note 26    See e.g. Preston-Jones v. Preston-Jones [1951] A.C. 391 per Lord Simonds.[Back]

Note 27    The trial judge may even have stated that he is satisfied on the balance of probabilities that the husband is not the father; see e.g. Watson v. Watson n. 20 supra.[Back]

Note 28    Until very recently under the law of Scotland the subsequent marriage of its natural parents did not legitimate a child if at the date of its conception or birth either of them was married to a third person. The Scottish Law Commission in a Report published in April 1967 (Cmnd. 3223: "Reform of the Law Relating to Legitimation per subsequens matrimonium") proposed the abolition of this rule and the Legitimation (Scotland) Act 1968 .which received the Royal Assent on 8th May, does this.[Back]

Note 29    14th September 1966 in "The Law in Action" series in the B.B.C. Third Programme; see The Listener, 6th October 1966, p. 493.[Back]

Note 30    We do not, however, suggest that corroborative evidence should be required to rebut the presumption of legitimacy.[Back]

Note 31    See para. 10 supra. [Back]

Note 32    [1931] N.Z.L.R. 559.[Back]

Note 33    Recently F. v. F. [1968] 2 W.L.R. 190 has provided another example of the difficulties which can be caused by applying the criminal standard of proof to the presumption of legitimacy. In this case two courts have now held that the child concerned was not the husband's and one court (whose view prevailed for five years) that it was.[Back]

Note 34    Though not a criminal offence in England it is (at any rate in theory) in a great many common law jurisdictions.[Back]

Note 35    [1948] P. 179.[Back]

Note 36   Preston-Jones v. Preston-Jones [1951] A.C. 391.[Back]

Note 37    In the same year the High Court of Australia refused to follow Ginesi v. Ginesi: Wright v. Wright (1948) Aust. L.J. 534.[Back]

Note 38    [1966] A.C. 643.[Back]

Note 39    See per Lord Denning at 669 and per Lord Pearce at 673.[Back]

Note 40    Rayden on Divorce, 10th ed., 1967, p. 176.[Back]

Note 41    In Re L. [1968] P. 119, Joyce v. Joyce [1966] P. 84, Burke v. Burke (1966) 110 Sol.J. 109.[Back]

Note 42    The concept of a surgical operation constituting an assault may well be out of date. See Professor Daube's views in 1966, Ethics in Medical Progress, a Ciba Foundation Symposium. Even so, taking a sample of blood in order to carry out tests to determine paternity would seem rightly, in the absence of consent, to amount to an assault, for, unlike a surgical operation the aim is not therapeutic.[Back]

Note 43    Latter v. Braddell (1881) 50 L.J.Q.B. 448.[Back]

Note 44    [1957] 1 W.L.R. 582.[Back]

Note 45    See Lanham, "Blood Tests and the Law", (1966) 6 Medicine Science and the Law, p. 190.[Back]

Note 46    Reported as a footnote to S. v. S. [1964] 3 All E.R. 915 at 917.[Back]

Note 47    In evidence to the Committee on the Age of Majority under the Chairmanship of Mr. Justice Latey: See Cmnd. 3342, para. 480.[Back]

Note 48    See n. 45 supra.[Back]

Note 49    Cmnd. 3342.[Back]

Note 50    Para. 474.[Back]

Note 51    Recommendation on page 118 of the Report.[Back]

Note 52    Para. 481.[Back]

Note 53    3 Curt. 1 at 14.[Back]

Note 54    S. v. B. (1905) 21 T.L.R. 219.[Back]

Note 55    See W. v. W. (N. 4) [1964] P. 67.[Back]

Note 56    R< v. Blakeman, Leeds Assizes, 20th March 1929.[Back]

Note 57    See Bartholomew, "The Nature and Use of Blood Group Evidence" (1961) 24 M.L.R. 313, n. 3.[Back]

Note 58    See n. 53 supra.[Back]

Note 59    [1968] P. 119.[Back]

Note 60    [1968] P. 119. (The hearings at first instance and on appeal are reported together.)[Back]

Note 61    This jurisdiction is derived from the Lord Chancellor's parental functions discharged on behalf of the Sovereign as parens patriae.[Back]

Note 62    n. 55 supra.[Back]

Note 63    L. v. L., The Times, 15th March 1968. 112 Sol. J. 294.[Back]

Note 64    The Times, 13th February 1968.[Back]

Note 65    [1968] 2 All E.R. 1023.[Back]

Note 66    [1968] P. at 171.[Back]

Note 67    Per Lord Denning M. R. at 1025.[Back]

Note 68    See n.24 supra.[Back]

Note 69    W. v. W. {No. 4) [1964] P. 67 and see para. 25 supra.[Back]

Note 70    See B.R.B. v. J.B. [1968] 2 All E.R. per Lord Denning M. R. at 1025. An undefended divorce petition can be transferred from the county court to the High Court and this could clearly be done to enable blood tests to be ordered, but this is an unnecessarily cumbersome and expensive way of solving the problem.[Back]

Note 71    [1966] 1 W.L.R. 187.[Back]

Note 72   At 188.[Back]

Note 73    19th July 1966, unreported.[Back]

Note 74    Quoted from the transcript.[Back]

Note 75    In our Working Paper we made several suggestions which we thought might be controversial and not necessarily desirable, in order to obtain comments on as wide a range of problems connected with blood tests as possible. As a result of our consultations, our final proposals differ, in some respects, from those canvassed in our Working Paper.[Back]

Note 76    The presumption of legitimacy applies even where a husband and wife are living apart during the possible period of conception. If, however, they are living apart under a valid decree of judicial separation or separation order and it is clear that the child must have been conceived after the decree or order was made, there is a presumption that the child is illegitimate.[Back]

Note 77    See paras. 58-65 infra.[Back]

Note 78    See n. 6 supra.[Back]

Note 80   Scandinavian Studies in Law, 1959 (published by Almqvist & Wiksell, Stockholm) at p. 16.[Back]

Note 81    [1968] P. 119.[Back]

Note 82    Ibid, at 159.[Back]

Note 83    Para. 36.[Back]

Note 84    S.2(l) provides: "... if it is proved that the accused . . . refused to consent to the taking of [a blood sample] ... his refusal may, unless reasonable cause therefor is shown, be treated as supporting any evidence given on behalf of the prosecution, or as rebutting any evidence given on behalf of the defence, with respect to his condition at that time" (emphasis added).[Back]

Note 85    S.2(3) provides: "A person who, without reasonable excuse, fails to provide a specimen of breath for a breath test. . ." (emphasis added).[Back]

Note 86    See e.g. Re L [1968] P. 119.[Back]

Note 87    See Affiliation Proceedings Act 1957, s.7.[Back]

Note 88    See paras. 19 to 21 supra.[Back]

Note 88    See para. 38 supra.[Back]

Note 90    [1968] P. 119.[Back]

Note 91    See F. v. F. [1968] 2 W.L.R. 190 per Rees J. at 193.[Back]

Note 92    We deal with the question of costs in proceedings other than affiliation in para. 66 infra.[Back]

Note 93    See para. 32 supra.[Back]

Note 94    Para. 43.[Back]

Note 95    Matrimonial Causes Rules 1968, r.31 (2).[Back]

Note 96    Matrimonial Causes Rules 1957, r.24 (5) and (6), substituted in 1963.[Back]

Note 97    A false declaration would attract the usual penalties.[Back]

Note 98    See para. 56 supra.[Back]

Note 99   See para. 9 supra.[Back]

Note 100    We are greatly indebted to the doctors and other experts who have helped us in the preparation of the material in this Appendix (see n.4 supra).[Back]

Note 101    Race and Sanger, Blood Groups in Man, 4th edition (1962), and see F. v. F. [1968] 2 W.L.R. 190 at 197 where Rees J. refers to the evidence given by a serologist as to the possibility of mutation.[Back]

Note 102    See e.g. "Medicolegal Application of Blood Grouping Tests. A Report of the Committee on Medicolegal Problems of the American Medical Association", (1957) 164 J. Amer. Med. Assoc. 2036 and A. S. Wiener, "Blood Grouping Tests in Disputed Parentage", (1956) 3 Jo. Forensic Med. 139.[Back]

Note 103    See "Mg, A New Blood Group Antigen of the MNS System", (1958) 3 Vox Sanguinis 81-91.[Back]

Note 104    Modified from Race and Sanger, op. cit. (n.l supra).[Back]

Note 105    See Stocker v. Stocker [1966] 1 W.L.R. 190 where evidence of haptoglobin grouping was used.[Back]

Note 106    See (1964) 204. Nature, 742.[Back]

BAILII: Copyright Policy | Disclaimers | Privacy Policy | Feedback | Donate to BAILII