Defendant appeals after his conviction by a jury of rape, a violation of § 566.030, RSMo.Supp.1982. The court found defendant to be a persistent sexual offender and sentenced him to fifty years’ imprisonment. We affirm.
On July 19, 1982, defendant stopped at the Air Park motel in Mexico, Missouri. He had stayed at the motel in the past, but on this occasion informed the clerk that he would not be spending the night. While there, he observed the victim, a motel resident, as she was riding her mo-ped and later as she was swimming. He learned her name and that she was a guest at the motel.
Contrary to his earlier plans, defendant returned to the motel at approximately 3:30 a.m. on July 20, 1982. Defendant inquired whether the “blond chick” was staying at the motel. The clerk told him her room number and assigned him a room. Defendant signed the hotel registration under another name.
Shortly thereafter, the victim was awakened by a man holding his hand over her mouth. Although the room was unlighted, she could clearly see her intruder from light through the curtains. The man, naked, displayed a knife and threatened to cut her throat if she screamed. He then forced his victim to have intercourse. Once completed, the victim asked what would happen. Defendant picked up the knife and stated that he had no choice; she could identify him. The victim jumped from the bed and was able to escape from the room. She ran, naked, to the motel office.
Police were summoned and immediately went to the victim’s room. There they discovered that the bathroom window had *265been removed and a table was located beneath the window frame. The outside light bulb also had been unscrewed. Police returned to the office and obtained a description from the victim; the clerk suggested that defendant fit this description. The police roused defendant from his room and recovered a knife clearly visible from the doorway. The victim identified the defendant at a show-up held in the motel office. Her in-court identification was equally unequivocal. She also identified the knife found in defendant’s room as that wielded by the intruder.
Police escorted the victim to the police station and from there to the hospital. Subsequent analysis of the hospital tests detected semen in the victim’s vaginal smear.
In his first point on appeal, defendant contends that the court erred in excluding evidence that the victim had intercourse with her boyfriend on July 16, four days prior to the offense charged. Defendant would have this court believe that the challenged evidence squarely falls under an exception to the rape shield statute which allows evidence of prior sexual activity to show an alternative source of semen. § 491.015.1(2), RSMo.1978. With this evidence, defendant, therefore, sought to explain away the presence of semen found in the victim’s vaginal smear.
The evidence belies defendant’s claim. At an in-camera hearing held pursuant to § 491.015.8, RSMo.1978, the physician who conducted the test stated that sexual relations three to four days prior to the test could not account for the presence of semen in the victim’s vaginal smear. This evidence, therefore, could have no reasonable bearing on the issues in defendant’s case. The evidence was properly excluded since “[introduction of such evidence serves only to humiliate and embarrass the witness in a public ‘fishing expedition’ that puts the complainant on trial instead of the appellant.” State v. Brown, 636 S.W.2d 929, 933 (Mo. banc 1982), cert. denied, 459 U.S. 1212, 103 S.Ct. 1207, 75 L.Ed.2d 448 (1983).
Defendant next challenges the admissibility of testimony by the state’s expert concerning the results of a laboratory analysis of the semen found in the victim’s vaginal smear. Defendant labels these results “speculative and prejudicial.” However, defendant made no objection to the introduction of this evidence at trial. Therefore, the matter is not properly preserved and our only review is under the plain error standard. Rule 30.20. Scientific tests and expert opinions rendered thereon are admissible if the principle involved is generally accepted in the scientific community. State v. Smith, 637 S.W.2d 232, 236 (Mo.App.1982). Moreover, the trial court has substantial discretion in the admissibility of expert testimony. Id. at 236; State v. White, 621 S.W.2d 287, 292 (Mo.1981). Applying these standards to the present case, we find no error, much less plain error, in the admission of the laboratory analysis.
The semen analysis test at issue is known as the Absorption Inhibition method. It allows an expert to group individuals into two categories: secretors and non-secretors. For secretors, the test determines the blood type of the assailant, as well as the victim, from such secretions as semen and saliva. Non-secretors make up twenty percent of the population and constitute an equally distinctive class whose blood type is not ascertainable from bodily secretions. See discussion in 12 Am.Jur. Proof of Facts “Identification of Seminal Fluids” § 12 at 321-22 (1961). The semen analysis in the present case determined that both the rapist and the victim were non-secretors; evidence also established that defendant is a non-secretor.
In the present case, the witness, a serologist expert, testified that the test at issue is accepted as accurate and reliable by the scientific community. Our review of the law bears out her claim. See 65 Am. Jur.2d, Rape § 61 (1972). The test has received tacit approval by the Missouri courts. See Starky v. State, 612 S.W.2d 861, 862 (Mo.App.1981) (counsel was not ineffective in failing to have this test conducted because in 1971, when the matter *266arose, the test was not in common usage and some evidence indicated the sample was contaminated); State v. White, 621 S.W.2d at 293-94 (Mo.1981) (no error in failing to order a co-defendant tested, not because the test was unreliable, but because in this particular case the serologist was unable to determine whether semen was produced by a secretor or non-se-cretor). The semen analysis test, furthermore, is widely regarded as both valid and reliable in other states.
In reaching its determination that the results of the test must be excluded as the test was not one which was accepted in the scientific community the trial court had before it little evidence to support that conclusion_ The defendant’s expert witness ... stated it was an accepted test. It appears to be a procedure regularly undertaken to determine blood type_ We conclude that on the evidence before it the trial court erred when it found that antigen testing was not generally accepted in the scientific community.
People v. McGee, 88 Ill.App.3d 447, 43 Ill.Dec. 641, 644, 410 N.E.2d 641, 644 (1980). See e.g. Nelson v. State, 247 Ga. 172, 274 S.E.2d 317, 320 (1981); State v. Williams, 229 Kan. 290, 623 P.2d 1334, 1340 (1981); People v. Nation, 26 Cal.3d 169, 161 Cal.Rptr. 299, 604 P.2d 1051, 1054-55 (1980); State v. Curtis, 295 N.W.2d 253, 256 (Minn.1980); State v. Pendergrass, 615 P.2d 201, 203 (Mont.1980); State v. Gray, 292 N.C. 270, 233 S.E.2d 905, 914-915 (1977).
We conclude the results in the present case were properly admitted by the trial court since semen analysis is widely accepted by both the scientific community and the courts. This evidence was relevant to establish that defendant is a member of that class to which the rapist belonged, namely, non-secretors. To this extent, the semen analysis, like blood tests and hair analysis, the reliability and admissibility of which are not questioned in Missouri, State v. Bolder, 635 S.W.2d 673, 688 n. 12 (Mo. banc 1982), cert. denied, 459 U.S. 1137, 103 S.Ct. 770, 74 L.Ed.2d 983 (1983); State v. Jones, 518 S.W.2d 304, 309 (Mo.App.1975) (blood typing); State v. White, 621 S.W.2d at 293; State v. Smith, 637 S.W.2d at 234-35 (hair analysis), was probative of a material issue in the case.
Defendant makes numerous other complaints. Among them, he challenges, pro se, the persistent offender statute, § 558.016, RSMo.1978. We have examined his pro se contentions and find them without merit. Defendant’s remaining complaints allege ineffective assistance of trial counsel. Many of these contentions were raised at the hearing on defendant’s motion for new trial. “Generally, the issue of ineffective assistance of counsel will not be reviewed on direct appeal. The issue can best be explored and disclosure of the pertinent facts can be obtained in a 27.26 proceeding.” State v. Sanders, 661 S.W.2d 52, 53 (Mo.App.1983).
We are not convinced that trial counsel failed to exercise that degree of skill and diligence of a reasonably competent attorney performing under similar circumstances and that defendant was thereby prejudiced. State v. Inman, 657 S.W.2d 395, 398 (Mo.App.1983). However, we conclude that this is not an “exceptional case” warranting review on direct appeal and therefore that defendant’s allegations are best left to a Rule 27.26 action.
Affirmed.
DOWD, C.J., and CRIST and PUDLOW-SKI, JJ., concur.