At 2:30 a.m., June 23, 1997, a patrolman of the Missouri State Highway Patrol saw a motor vehicle exceeding the speed limit in a “construction zone.” The patrolman stopped the vehicle; its operator was Patrick J. Rhodes (“Driver”).
At the patrolman’s request, Driver performed six “field sobriety tests.” Driver’s performance, coupled with the result of a “portable breath test” administered by the patrolman, convinced the patrolman that Driver was intoxicated. Accordingly, the patrolman arrested Driver for driving while intoxicated at 2:42 a.m. and took him to the local sheriffs office where, at 3:27 a.m., the patrolman, using a “BAC Data-Master,” administered a “breath test” to Driver to determine the alcohol concentration in his blood. The test showed an alcohol concentration of .12 percent.
Pursuant to § 302.505,1 the Director of Revenue (“Director”) suspended Driver’s motor vehicle operator’s license. The suspension was upheld after administrative review, whereupon Driver filed a petition for trial de novo per § 302.535.
Following the presentation of evidence, the trial court found the patrolman had probable cause to arrest Driver for driving while intoxicated, but that the alcohol concentration in Driver’s blood “was not .10% or greater at the time he was operating his vehicle.” Consequently, the trial court set aside Director’s suspension of Driver’s license.
Director appeals. Director’s sole point relied on avers:
“The trial court erred in setting aside the suspension of Driver’s driving privileges under § 302.505 ... because the Director proved a prima facie case that Driver’s BAC was at least .10% in that the Director offered into evidence Driver’s breath test result showing his blood alcohol content was .12% and Driver did not rebut the Director’s prima facie case by showing that the blood alcohol content was actually lower than .10%.”
To suspend Driver’s license, Director was required to demonstrate, by a preponderance of the evidence, that (1) Driver was arrested upon probable cause to believe he was driving while intoxicated, and (2) the alcohol concentration in Driver’s blood was in fact ten-hundredths of one percent or more by weight at the time he was driving. Halmich v. Director of Revenue, 967 S.W.2d 693, 695[1] (Mo.App. E.D.1998); Whitworth v. Director of Reve*599nue, 953 S.W.2d 142, 143[1] (Mo.App. E.D.1997); Kienzle v. Director of Revenue, 944 S.W.2d 326, 327[2] (Mo.App. S.D.1997); Brandom v. Director of Revenue, 931 S.W.2d 510, 511[1] (Mo.App. S.D.1996). Inasmuch as the trial court found Director satisfied the first element, only the second element is in issue in this appeal.
Director makes a prima facie case on the second element if Director establishes by credible and competent evidence that proper chemical analysis showed a motor vehicle operator’s blood alcohol concentration was ten-hundredths of one percent or more by weight.2 Andersen v. Director of Revenue, 944 S.W.2d 222, 223 (Mo.App. W.D.1997), citing Collins v. Director of Revenue, 691 S.W.2d 246, 252 (Mo. banc 1985). Once Director presents a prima facie case, the burden shifts to the motor vehicle operator to establish by a preponderance of the evidence that his blood alcohol concentration was less than ten-hundredths of one percent by weight at the time he was driving. Green v. Director of Revenue, 961 S.W.2d 936, 938[3] (Mo.App. E.D.1998); Andersen, 944 S.W.2d at 223-24.
Driver testified he consumed no alcoholic beverage on June 22, 1997, until “between 9:30 and 10:00” that evening. At that time, avowed Driver, he arrived at Rock Lane Lodge and began drinking a can of beer. Driver testified he drank two more cans of beer at that site before departing “around 12:30 or 1:00.”
According to Driver, he drove “out to the lake,” a journey that required some “30, 45, minutes.” There, at a marina, Driver consumed a “drink” consisting of “Jim Beam and Diet Coke.” Driver related that he finished the drink “probably like right at 2:20 [a.m.],” then left the marina “to go home.” As reported in the first paragraph of this opinion, the patrolman stopped Driver at 2:30 a.m.
A companion who accompanied Driver to Rock Lane Lodge and remained with him until the patrolman stopped him corroborated Driver’s account of his drinking.
A forensic chemist, called as a witness by Driver, testified he (the chemist) calculated Driver’s blood alcohol concentration at the time the patrolman stopped him. The chemist performed the calculation by applying “the Widmark equation”3 to Driver’s “rate of drinking” set forth earlier in this opinion.
According to the chemist, alcohol entering the body by mouth is “absorbed into the bloodstream intact” and eventually eliminated. The chemist said: “[I]f an individual drank all the liquor early in the drinking episode, it wouldn’t have the tremendous effect in a BAC as it would if the individual drank the bulk of the liquor towards the end of it. That’s why I ... ask[ed] for specific times for specific drinks.”
Without objection by Director, the chemist testified “with a reasonable degree of ... scientific certainty” that his “estimation” was that Driver “would have a BAC between .032 and .055 at the time of initial contact, which was 2:30 a.m.” The chemist continued:
“I used the value of peak BAC as achieved approximately one hour after cessation of the drinking period.... He quit drinking ... at 2:20. The BAC was run at 3:27. The initial contact was at 2:30, ten minutes after the last drink, so it is my opinion that at 2:30 his BAC would be ... inclining. He would still be absorbing alcohol. At approximately ... 3:20, it would be my opinion that this could represent the peak BAC.... *600[T]he bulk of the [Jim Beam and Diet Coke] may have been consumed closer to ... 2:20. Then that amount of alcohol ... which I ... calculated to be approximately .027 percent, would not be in his system or would just be beginning to enter his system at the time of the initial contact.”
On cross-examination, the chemist admitted Driver’s blood alcohol concentration at the time he was stopped by the patrolman “could have been above or below” ten-hundredths of one percent.
At the conclusion of the evidence, the trial court announced that based on the undisputed testimony of the chemist, the court found Director failed to prove Driver’s blood alcohol concentration was ten-hundredths of one percent or more by weight at the time he was driving.
This court must affirm the trial court’s judgment unless there is no substantial evidence to support it, unless it is against the weight of the evidence, unless it erroneously declares the law, or unless it erroneously applies the law. Hawk v. Director of Revenue, 943 S.W.2d 18, 19-20[1] (Mo.App. S.D.1997). In determining whether there is substantial evidence to support the judgment, this court must defer to the trial court on the issue of credibility of witnesses, id. at 20[2], as it was the trial court’s prerogative to accept or reject all, part, or none of any witness’s testimony. Id. at [5]. Accordingly, this court views the evidence in the light most favorable to the trial court’s judgment, id. at [3]. even if there is evidence that would support a different result. Id. at [4].
Director maintains the chemist’s testimony was insufficient to rebut Director’s prima facie case on the second element (set forth earlier in this opinion) in that the chemist “only speculated” as to what Driver’s blood alcohol concentration was at the time he was driving. According to Director, the instant case is governed by Green, 961 S.W.2d at 936, which, says Director, presents the identical issue.
In Green, Director suspended a motor vehicle operator’s license for the same reason Director suspended Driver’s license in the instant case. Id. at 937. The trial court, following trial de novo, reinstated the license in Green, whereupon Director appealed. Id.
Evidence in Green favorable to the trial court’s ruling was that the motorist arrived at a bar at 11:30 p.m., drank two beers, then “chugged” the last half of his girlfriend’s beer about five minutes before he was stopped at 12:37 a.m. Id. at 938. A BAC DataMaster test administered at 1:08 a.m. showed a blood alcohol concentration of .104 percent by weight. Id. at 937-38.
An “expert witness on toxicology and pharmacology” testified in Green that alcohol reaches its peak in a person’s bloodstream between one-half and two hours after ingestion. Id. The average for beer, said the expert, is 52 minutes. Id. at 938. The expert opined that the motorist’s blood alcohol concentration was rising at the time the test was administered. Id.
The appellate court held the motorist’s evidence in Green insufficient to rebut Director’s prima facie case. Id. at 939. The court emphasized that the expert conceded the motorist’s blood alcohol concentration at the moment he was driving could have been lower or higher than the .104 percent recorded 31 minutes after he was stopped. Id. at 938. Additionally, said the court, the expert’s theory demonstrated the motorist’s blood alcohol concentration could have been falling at the time of the test, as “peaking” could have occurred as quickly as 30 minutes after ingestion because the motorist had eaten nothing since breakfast. Id.
In the instant case, Driver endeavors to distinguish Green by pointing out that the expert there did not render his opinion within a reasonable degree of scientific certainty,4 whereas the chemist in the in*601stant case testified within a reasonable degree of scientific certainty that Driver’s blood alcohol concentration would have been between .032 percent and .055 percent when he was stopped. However, a close reading of the chemist’s testimony in the instant case reveals he qualified his answer by characterizing it as an “estimation.” Furthermore, like the expert in Green, the chemist in the instant case acknowledged Driver’s blood alcohol concentration at the time he was stopped could have been above ten-hundredths of one percent. Finally, the chemist in the instant case calculated that the Jim Beam and Diet Coke had an “alcohol absorption rate” of .027 percent. If one accepts the chemist’s theory that none of the alcohol from that drink had been absorbed into Driver’s bloodstream when he was stopped, and further accepts the chemist’s theory that all of the alcohol from that drink had been absorbed into Driver’s bloodstream at the time of the BAC Data-Master test, the .027, combined with .055 (the maximum blood alcohol concentration attributed by the chemist to Driver at the time he was stopped) amounts to only .082 percent, a figure that falls short of the .12 percent test result — the difference is obviously .038 percent.
Director’s lawyer pointed out this anomaly to the chemist during cross-examination and asked how, if the chemist’s theory was correct, Driver would have registered .12 percent on the BAC DataMaster. The chemist’s only explanation was that Driver’s “individual tolerances, absorption or elimination rates” could account for the .12 percent reading. Significantly, the chemist added: “So I cannot state that it ... would definitely be below .1.”
Having carefully compared the evidence in the instant case with the evidence in Green, this court holds the evidence favorable to Driver in the instant case is weaker than the evidence favorable to the motorist in Green.
First, the BAC DataMaster in the instant case produced a reading of .12 percent, whereas the reading in Green was only .104 percent. Consequently, Driver’s blood alcohol concentration at the time of the test was .016 percent higher than the motorist’s blood alcohol concentration in Green. The motorist’s blood alcohol concentration in Green would have had to be only .005 percent lower at the time he was stopped than at the time of the test in order to be beneath .10 percent at the time of the stop. Yet, Green held the evidence insufficient to rebut Director’s prima facie case.
In the instant case, Driver’s blood alcohol concentration would have had to be .021 percent lower at the time he was stopped than at the time of the test in order to be beneath .10 percent at the time of the stop. That is four times the amount in Green (.005). It is thus evident that Driver in the instant case faced a more arduous challenge in rebutting Director’s prima facie case than the motorist in Green.
Second, the chemist’s theory about Driver’s alcohol absorption rate in the instant case would have produced, at most, a reading of only .082 percent at the time Driver took the BAC DataMaster test. Yet, the result of that test was .12 percent. Driver does not assail the accuracy of that reading. Thus, the chemist’s testimony in the instant case is inherently irreconcilable.
Finally, like the expert in Green, the chemist in the instant case admitted Driver’s blood alcohol concentration at the time he was stopped could have been .10 percent or more.
This court therefore holds as a matter of law that the evidence in the instant case, even viewed most favorably toward Driver, is insufficient to rebut Director’s prima *602facie case on the second element, i.e., the alcohol concentration in Driver’s blood was ten-hundredths of one percent or more by weight at the time he was driving.
The judgment is reversed and the case is remanded to the trial court with a directive to enter judgment affirming Director’s suspension of Driver’s motor vehicle operator’s license.
PREWITT, P.J., and PARRISH, J., concur.. References to statutes are to RSMo Cum. Supp.1996.
. Director presented uncontradicted evidence that the BAC DataMaster analyzes a motor vehicle operator’s "breath sample” to determine his or her blood alcohol concentration. The instrument, of course, reveals the blood alcohol concentration at the time the motor vehicle operator provides the sample, not at the time he or she was driving.
. The chemist explained that the Widmark equation "was devised back in the 1930s by a German scientist.”
. The expert in Green testified the breath test result was not accurate within a reasonable *601degree of scientific certainty. Id. at 938. However, nothing in Green indicates the expert testified his theory about alcohol absorption was accurate within a reasonable degree of scientific certainty.