Two issues are raised in this litigation, the first concerning a determination of the paternity of a female child born on January 4, 1971, to Linda Kathleen Richardson, appellee herein, and the second relating to the custody of Julie Ann Richardson, born November 1, 1969. On trial, the court held that William Glenn Richardson, appellant herein and husband of appellee, was the father of the daughter born in January 1971, and held that appellee should also have custody of the older child, Julie Ann, admittedly the daughter of the parties, with the right of visitation to the appellant of three weeks each third month, such visitation to com*245menee on July 1, 1971. The court directed that appellant pay $15.00 per week for the support of the children. From the decree so entered, appellant brings this appeal.
The background relating to the first issue is found in paragraph 7 of the court’s decree, which reads as follows:
“On August 18, 1970, upon the trial of this cause, this plaintiff and corss-defendant [appellee] was pregnant. Therefore, this court took this case under advisement to allow both parties to receive professional counseling and to allow for the birth of this plaintiff’s child. Further, on August 18, 1970, this court ordered plaintiff, defendant, and said minor child to have a blood test taken as soon as it was reasonable to do so. The court finds that on the 9th day of February, 1971, both plaintiff and defendant consented to go together to the laboratory of Dr. Annette Landrum to have the necessary blood test completed as ordered by the court to determine if this defendant could be excluded as the father of said child born on January 4, 1971. Thereafter, the required blood tests were performed by Dr. Annette Landrum, M.D., F.C.A.P., Pathologist, and report submitted to this court dated February 10, 1971, stating, ‘We are able to exclude Glenn Richardson as the father of the second daughter 1 on the basis of the S Factor. Since Glenn Richardson and Linda Richardson are both negative for the S Factor and since the second daughter is positive for the S Factor, she would have to have inherited it from some other source. Therefore, Glenn Richardson could not possibly be the father of the second daughter.’ Upon the request of defendant’s [appellant’s] attorney and without objection from plaintiff or her attorney, this report from Dr. Annette Landrum dated February 10, 1971, was received into evidence in this cause in accordance with Ark. Stats. § 34-705.1. The court further finds that plaintiff and her attorney were given the opportunity to cross-examine Dr. Annette Landrum with regard to her professional qualifications *246as well as the integrity of said blood test. Plaintiff and her counsel waived the opportunity to cross-examine Dr. Annette Landrum in either respect.”
However, the court found that the parties lived together until about May 15, 1970; that the divorce action was filed on behalf of Mrs. Richardson on June 16, 1970, and at the time of the hearing on August 18, 1970, appellee was five months pregnant. It was further found that appellant made no claim or offered no proof of impotency or non-access to appellee “to overcome the strong presumption of legitimacy and paternity of said minor child, and this court, therefore, finds that this defendant and cross-complainant is the father of the second daughter born on January 4, 1971. This court finds that this defendant and cross-complainant is as a matter of fact and as a matter of law the father of the newly born child”.
Though with natural and understandable reluctance, we are compelled to find that the court erred in its determination that appellant was the father of the second child. The trend over the nation within the last several years has been to recognize the value and accuracy of blood tests to determine paternity. In Wigmore on Evidence, section 165a p. 610, it is stated:
“In one specific biological trait, viz. blood-groups, scientific opinion is now in accord in accepting the fact that there is a causative relation between the trait of the progenitor and the trait of the progeny. In other words, the blood-composition of a child may be some evidence as to the child’s paternity. But thus far this trait (in the present state of scientific discovery as generally accepted) can be used only negatively, i.e., to evidence that a particular man P is not the father of a particular child C.”
In 9 Uniform Laws Annotated, p. 102, there is a discussion on the subject of “Blood tests to disprove Paternity”, in which it is stated:
“In paternity proceedings, divorce actions and other types of cases in which the legitimacy of a child is in *247issue, the modern developments of science have made it possible to determine with certainty in a large number of cases that one charged with being the father of a child could not be. Scientific methods may determine that one is not the father of the child by the analysis of blood samples taken from the mother, the child, and the alleged father in many cases, but it cannot be shown that a man is the father of the child. If the negative fact is established it is evident that there is a great miscarriage of justice to permit juries to hold on the basis of oral testimony, passion or sympathy, that the person charged is the father and is responsible for the support of the child and other incidents of paternity. *** The conclusion should be final if there is no dispute among the experts. There is no need for a dispute among the experts, and true experts will not disagree. Every test will show the same results.”
In the New York case of Anonymous v. Anonymous, 150 N.Y.S. 2d 344 (Supreme Court, Appellate Division, Second Department), the court held that even though the husband had lived with the wife for five years after the birth of his putative children (twins) before discovering evidence of adultery the husband was entitled to blood grouping tests of himself, his wife, and the children, for purposes of excluding his paternity of these children. This holding was rendered despite the fact that the husband had lived with the wife during the gestation period and for the five years mentioned. A discovery of correspondence from a third party to the wife contained significant remarks to indicate intimate relations between the wife and the writer. This precipitated the action for divorce. It was argued by the wife that the husband and wife had continuously lived together during the period of gestation of the twins and for more than five years afterward, and that the husband had accepted the children as his own and had never questioned his paternity. The wife stated that the letters were a manufactured fabrication. The court commented:
“Reason and logic, as well as a recognition of the modern advances in science, compel a determination *248that the presumption of legitimacy is not conclusive but rebuttable. The probative value of the results of skillfully conducted blood grouping tests has been widely accepted. The tests of course will be relevant only if they show noncompatibility as between the blood of defendant, the plaintiff, and the twins. If so, such evidence should be deemed conclusive as to nonpaternity. ***
There is no doubt that with the passing of years and the advance of science the age-old concept has gradually given way to the sway of reason, and that the presumption of legitimacy has been withering and shrinking in the face of scientific advances, (citing cases) Presumptions are looked upon as the bats of the law, flitting in the twilight, but disappearing in the sunshine of actual facts.’ (citing cases) It cannot be gainsaid that we have now reached the point where presumptions must yield to modern scientific facts.”
In the Nebraska case of Houghton v. Houghton, 137 NW 2d 861 (1965), the wife sued the husband for divorce. After filing her petition, she filed a supplementary petition, alleging that the parties had resumed the marriage relationship and were expecting the birth of a second child, but acts of cruelty had again occurred, and she therefore sought a divorce. The husband answered, admitting he was the father of the first-born child, and also admitting that he had subsequently had sexual relations with his wife, but he denied being the father of the expected child. This child was born on October 1, 1963, and the trial court found the husband to be the father and awarded a decree of divorce to the mother, along with the custody of the two minor children. The husband appealed, alleging inter alia that the trial court had erred in its finding that he was the father of this second child; further, that the court had erred in not finding that the results of blood tests of the mother, husband and the infant were conclusive in showing that he could not be the father; that such evidence overcame the presumption that the child was his. The Supreme Court of Nebraska said:
“We have reviewed the evidence of the doctor carefully and we think the qualifications of his technicians *249to make the tests have been adequately shown as have his qualifications to interpret them. Although this is the doctor’s first test for nonparentage, other tests of that nature have been made by his partner using the same facilities. The plaintiff introduced no medical testimony to refute them. The doctor who performed the tests was agreed upon by both parties and was not working for either one of them. We conclude there is nothing in the record which would indicate any defect in the testing methods, and his testimony and conclusions remain unshaken.
Having determined that no defect in the testing methods appear from the evidence concerning the blood tests in the case before us, the next question presented is what consideration and weight should be given to the results disclosed by them. Various statutes have been passed providing for blood tests in cases where paternity is an issue. Plaintiff contends that in the absence of a statute specifically authorizing such blood tests this court should not consider them in the present case at all. The Supreme Court of New Hampshire, in Groulx v. Groulx, 98 N.H. 481, 103 A. 2d 188, 46 A.L.R. 2d 994, a case which had been decided in the trial court before the passage in that state of such a statute, held that judicial recognition should be accorded the accuracy and reliability of blood grouping tests to disprove paternity. See, also, State v. Damm, 62 S.D. 123, 252 N.W. 7, 104 A.L.R. 441, where the court corrected and clarified its original holding of 3 years before, stating: ‘We therefore say, without further elaboration or discussion, that it is our considered opinion that the reliability of the blood test is definitely, and indeed unanimously, established as a matter of expert scientific opinion entertained by authorities in the field, and we think the time has undoubtedly arrived when the results of such tests, made by competent persons and properly offered in evidence, should be deemed admissibile in a court of justice whenever paternity is in issue.’ Many authorities are thereafter cited. We hold this court should likewise take judicial notice of the scientific accuracy and reliability of such tests.”
*250In the Colorado case of Beck v. Beck, 584 P. 2d 731 (1963), the evidence was conflicting as to whether the husband had access to the wife during the possible period of conception. The trial court directed a verdict in favor of the wife and the husband filed a motion for a judgment non obstante veredicto attaching thereto a blood test taken by agreement, said blood test finding that the child could not be the som of the husband because the child possessed antigen “E”, a factor or element not possessed by either the husband or the wife. The trial court granted the husband’s motion, and the appeal followed. The Supreme Court, in an opinion by Justice Sutton, affirmed the action of the trial court in granting the motion non obstante veredicto stating:
“In the instant case the accuracy of the blood test was not challenged by the wife. Her only thesis on this issue being that it is incompetent to overcome the presumption of legitimacy. We must hold otherwise. To hold as she contends would be contrary to an established scientific fact. In this connection see the excellent dissent in Prochnow v. Prochnow, 274 Wis. 491, 80 N.W. 2d 278 (1955) with which we agree and which contains a thorough discussion of this type of medical evidence. We note that following the Prochnow decision, and undoubtedly due to the persuasive dissent therein, the Wisconsin statute was amended to make it mandatory for the court to order the blood test when relevant in a paternity action and that the result shall be conclusive where exclusion is established.”
In Commonwealth v. D’Avella, 162 N.E. 2d 19, the Supreme Judicial Court of Massachusetts said:
“Here we have definite evidence based on blood tests, the integrity of which is conceded, excluding the defendant. To hold in the light of such evidence that it was still open to the trier of fact to find that the defendant was the father would be egregiously unrealistic. The reliability of such tests to prove nonpaternity is well established as a scientific fact. Evidence which is regarded and acted upon every day as con-*251elusive by skilled scientists outside of court ought not to be treated merely as some evidence (to be believed or disbelieved as the trier of fact sees fit) when it is adduced in court. We cannot close our eyes to the overwhelming weight of scientific opinion on this subject and we take judicial notice of it.”
Numerous other cases from other jurisdictions hold similarly.
The Arkansas General Assembly has recognized this trend, and in 1955 passed an Act which is presently codified as Ark. Stat. Ann. § 34-705.1 (Repl. 1962) et seq. 2 This section is included in the chapter on “Bastardy” and appellee argues that it only relates to bastardy proceedings and has no application to divorce actions (it is not explained why the words “husband” and “wife” are used). Such argument is really of no moment here, for scientific advances would support the offering of this evidence, even though no statute authorized it.
Some jurisdictions have held the blood test excluding paternity to be conclusive, while others have only considered it as strong evidence, to be considered in conjunction with other evidence. In the case before us, it is not necessary that we determine whether the results of these tests are always conclusive, for, together with the other proof in this case, we think the result establishes conclusively that appellant is not the father of the second daughter. While Mrs. Richardson denied acts of adultery, the weight of evidence is contrary to her testimony. No good point *252would be served in detailing this evidence; it is enough to say that it is convincing, and when the blood grouping tests of Dr. Landrum are added to it, the evidence is certain and conclusive.
Let it be remembered that both husband and wife consented to go to the laboratory of Dr. Landrum, together with the baby, to have the necessary blood tests completed, and, though given the opportunity to cross-examine Dr. Landrum with regard to her qualifications and the integrity of the blood tests, appellee declined to do so. No question whatsoever has been raised about the competence of the doctor to conduct the tests, nor has any objection been made to the manner in which is was conducted. As stated, we hold that it has been established that William Glenn Richardson is not the father of the second daughter.3
Despite this holding, we are of the view that the chancellor’s order as to custody should not be disturbed. Appellee lives with her parents, who appear to be substantial citizens of Fort Smith, and there is no evidence of misconduct since she has been living with the parents. Appellee does not work, and is accordingly able to look after the children, and the facilities for taking care óf them are more than adequate. The husband lives with his parents in White House, Tennessee, and of course, his mother would have to look after the child, while appellant is at work. Perhaps our view is somewhat influenced by the fact that we are hesitant to separate these two little girls. Should that become necessary, the chancellor has full authority to make any changes in his order of custody, and, in fact, the chancellor’s order finds that the children should be awarded to appellee “at this time for so long as said plaintiff and cross-defendant continues to reside in the home of her parents.”
Nor do we deem it proper to change the amount of money ordered for support of the children. The court di*253rected appellant to pay $15.00 per week, and, though appellant is not liable for the support of the second daughter, we certainly cannot find that the $15.00 per week is in excess of the amount required for the support and maintenance of the older daughter, Julie Ann.
In accordance with what has been said, the decree is reversed as to the first point, and the cause remanded with directions to enter a decree not inconsistent with this opinion.
Fogleman, J., would reduce the amount for child support.