OPINION OF THE COURT
Kevin C. Fogarty, J.This paternity proceeding was instituted by a guardian ad litem to determine which of two respondents is the father of the child. The child’s mother died prior to the institution of the proceeding and the action was commenced on behalf of the child. (See Matter of Alicia C. v Evaristo G., 114 Misc 2d 764.) In the absence of the mother, direct proof is difficult to obtain. Here both respondents claim to be the father.
A composite human leucocyte antigen (HLA) blood test was ordered. Dr. Leon Sussman, a leading expert in the field, performed the tests and testified as to the procedures used in testing and as to the interpretation of the test results in order to lay a foundation for its admissibility into evidence. (Matter of Rosemary W. v Bruce A., 113 Misc 2d *565745; but see L 1982, ch 695, amdg Family Ct Act, §§ 418, 532; and CPLR 4518 on the admissibility of the HLA test into evidence.)
“Generally speaking, a predicate for the admission of expert testimony is that its subject matter involve information or questions beyond the ordinary knowledge and experience of the trier of the facts. Moreover, the expert should be possessed of the requisite skill, training, education, knowledge or experience from which it can be assumed that the information imparted or the opinion rendered is reliable * * *
“[T]he court still had to ascertain whether, [the doctor] * * * exhibited a degree of confidence in his conclusions sufficient to satisfy accepted standards of reliability.” (Matott v Ward, 48 NY2d 455, 459.)
“[T]he doctor intends to signify a probability supported by some rational basis.” (Miller v National Cabinet Co., 8 NY2d 277, 282.)
In his testimony, Dr. Sussman clearly established that the respondent Henry P. was excluded as a possible father on both the red blood cell testing and the HLA white cell testing.
In seeking however to establish the probability of parentage of the respondent Evaristo G., the following testimony was given:
“Q. Let me ask one question, Doctor. Nationalities and race are a factor in this?
“A. Yes * * * It certainly could be dependent upon racial characteristics. However, an exclusion of paternity is not dependent upon that. An inclusion is dependent upon classifying them as Caucasian or negroid. In this case they were classified as Caucasian.
“Q. Under the Caucasian classification, is that the figure upon which you based the 98.5?
“A. That’s correct.
“Q. Let’s assume for the sake of argument that they could be classified as negroid. Let’s assume that’s part of their background. Would that alter the percentages and the probability?
*566“A. Yes, it would alter it considerably.
“Q. In which way?
“A. If one were to use the black characteristics the probability of paternity would be 54 percent; considerably different.
“Q. The probability would be reduced to 54 percent?
“A. Yes; from 98 and a half depending upon whether one considered them Caucasian or negroid.
“Q. Did you interview or does anybody interview the parties before their blood is drawn?
“A. It’s our impression that we’re not permitted to. We have to draw a conclusion from the appearance of the people * * *
“Q. And you don’t make any inquiry as to that when you discuss the matter?
“A. No ***
“Q. Now, this question as to the probability depending on a racial background, if an alleged father is partially Caucasian and partially negroid would the probabilities also fall somewhere between the two numbers that you gave us?
“A. There is no way of judging that. We’re having our greatest difficulty with the Hispanic population, and the reason why is because the Hispanic population is very, very mixed. It consists of Indian, Spanish, black and white. So it’s impossible to get a single table that will give us a leeway in judging the frequency of certain factors in a group that is so mixed. So we arbitrarily pick a black or white standard and use it. You could not combine the two tables and come up with an honest calculation.”
Whether or not the respondent Evaristo G., a dark-skinned Hispanic, is Caucasian or black or a mixture of races is not of importance. What is important is that a test based upon an arbitrary, subjective determination of race by the examiner cannot be the basis for a scientific procedure which is to be recognized. (Matott v Ward, supra.)
“Report of the Expert * * * the report should be sufficiently detailed as to the findings and the expert’s opinion based on the findings as to minimize questions. If the test *567shows a strong likelihood of paternity * * * this evidence should be given to the court along with a description of the method used for calculating likelihood of paternity.” (Joint AMA — ABA Guidelines: Present Status of Serologic Testing in Problems of Disputed Parentage, 10 Family LQ 247, 282; emphasis supplied.)
The failure of the expert here to make any inquiry to determine the proper race (id., pp 267-275, gene frequencies of HLA A and B antigens by race and testing within various blood groupings to exclude or include certain races) precludes a finding by the court that the expert’s opinion is based on reliable information. A plausability of paternity of 98.5% (very likely) cannot be accepted when use of a possible different chart would produce a 54% probability.
Excluding, therefore, the HLA results from evidence as they apply to respondent G., there is simply an insufficient amount of evidence to establish by the clear, convincing and entirely satisfactory proof required that the respondent G. is the father of the child Z.
The records of New York Hospital reveal some dispute as to the period of gestation: the mother indicating 44 weeks, the doctors 39 weeks. It also shows that this was the deceased mother’s third pregnancy whereas the testimony of the respondent G. was directed solely to this pregnancy. While Mr. G. appeared in part of the records as the father, the original birth certificate shows no father and the mother later on signed a declaration that Henry P. was the father of Z.
The petitioner has presented members of the family of the respondent G. who have testified that he had a close personal relationship with the deceased mother and that they lived together at different times including the period prior to the pregnancy and for a period up to a year after the child was born on August 12, 1978. The testimony of Elsie R. and Irina G., the sisters of the respondent G., as well as that of Alicia C., the grandmother, and Barbara C., sister of deceased mother, was similar in most aspects. However, it differed in such particularly as the length of the periods that the deceased mother and the respondent G. were together and separated.
*568Mr. G. testified that he met Zulema C., the deceased mother, in the summer of 1973, kept company with her foi one year, and then moved into an apartment at 100 West 93rd Street in Manhattan where they resided for a year and a half. They then separated for two months and thereafter resumed living together on 89th Street between York and First Avenues for nearly a year. They separated for five months and resumed living together on 55th Street between Second and Third Avenues, separated again for a month and a half, and resumed living together at the home of Alicia C. in early 1977 where they lived for approximately 10 months. He learned of her pregnancy in December, 1977 and moved with her in the fifth month of her pregnancy to 205 West 105th Street where they lived together until they separated in August, 1979. Later that month, the respondent joined the United States Navy.
The Appellate Division, Second Department, has recently held:
“A critical element of proof necessary to sustain the petition is fatally absent in this case, viz., the date, or dates, upon which [the parties] had sexual intercourse capable of causing the conception of the subject child * * *
“Only vague and imprecise evidence of the dates of their sexual intercourse was presented, which was incapable of sustaining petitioner’s heavy burden of proof in this paternity proceeding, i.e., proof to the point of entire satisfaction by clear and convincing evidence.” (Matter of Commissioner of Social Servs. [Patricia A.] v Philip DeG., 88 AD2d 911, 912.)
Additionally, the respondent G. testified that he was present in the delivery room for the birth of his son and yet the hospital records show that he was not present.
Considering all of the evidence and the lack of probative value that can be given to the HLA test in this proceeding the court finds that the petition has not been established by clear and convincing proof that is entirely satisfactory. The matter of custody and guardianship of the child is therefore set down for a hearing on September 30, 1982 in Part IV.
So ordered.