OPINION
DUGGAN, Justice.The jury found appellant, Elija T. Smith, guilty of possession of less than 28 grams of cocaine. The trial court found two enhancement paragraphs true and assessed punishment at 30-years confinement. In his sole point of error, appellant argues that the evidence was insufficient to prove possession of cocaine. We affirm.
On May 9, 1992, Charles E. Thompson, Sr. was asleep on the couch in his daughter’s house when someone entered, took his car keys from a coffee table, and stole his car. *721On more than one occasion, Thompson’s son-in-law, Gilbert Rodriguez, who also lived in the house, had stolen his vehicle. When he awoke, Thompson reported his car stolen.
At approximately 4:30 a.m., on May 10, 1992, Officer N.E. Malloy of the Harris County Sheriffs Department, saw a vehicle weaving in and out of the lane on a street in northwest Houston. He had the vehicle pull over; Rodriguez was the driver and appellant was the passenger. Because he determined that an arrest warrant had been issued for driving while his driver’s license was suspended, Malloy placed Rodriguez under arrest. When Rodriguez got out of the vehicle, Malloy saw a “chrome-looking tube, the kind they use for smoking crack cocaine,” on the floorboard of the driver’s side.
Malloy placed Rodriguez in the patrol car, and removed the pipe (Pipe One) from the floorboard. He then asked appellant to get out of the vehicle so that he could search it. Malloy found two more pipes (Pipes Two and Three), one made of metal, the other of glass; one was located on the floorboard of the passenger’s side, the other on the passenger’s side of the front seat. Malloy then arrested appellant. Malloy saw a substance in the glass pipe. Although it was not recorded in his police report, Malloy testified that he believed both Rodriguez and appellant were intoxicated on crack cocaine because they appeared fidgety and wide-eyed, and had accelerated pulse rates.
A field test of the substance in the two pipes located near appellant indicated a positive test for cocaine. A subsequent chemical analysis of the combined substance taken from three pipes revealed 7.06 milligrams of pure cocaine.
In reviewing the sufficiency of the evidence, an appellate court must view the evidence in the light most favorable to the verdict to determine if any rational trier of fact could have found the essential elements of the crime beyond a reasonable doubt. Jackson v. Virginia, 443 U.S. 307, 319, 99 S.Ct. 2781, 2789, 61 L.Ed.2d 660 (1979). This Court may not sit as a thirteenth juror and disregard or reweigh the evidence. Moreno v. State, 755 S.W.2d 866, 867 (Tex.Crim.App.1988). If there is evidence that establishes guilt beyond a reasonable doubt, and the trier of fact believes that evidence, we are not in a position to reverse the judgment on sufficiency of evidence grounds. Id.; Glass v. State, 761 S.W.2d 806, 807 (Tex.App.—Houston [1st Dist.] 1988, no pet.). The jury, as trier of fact, is the sole judge of the credibility of witnesses, Sharp v. State, 707 S.W.2d 611, 614 (Tex.Crim.App.1986), cert. denied, 488 U.S. 872, 109 S.Ct. 190, 102 L.Ed.2d 159 (1988), and may believe or disbelieve all or any part of a witness’ testimony. Id. at 614; Smith v. State, 789 S.W.2d 419, 420 (Tex.App.—Houston [1st Dist.] 1990, pet. ref d). A jury may believe a witness even though his testimony is contradicted. Sharp, 707 S.W.2d at 614.
Appellant contends that we can give no weight to Malloy’s testimony to prove that the substance possessed by appellant was cocaine. Over the objection of appellant’s trial counsel, Malloy testified that the field test of Pipes Two and Three revealed the presence of cocaine. Appellant cites Curtis v. State, 548 S.W.2d 57, 59 (Tex.Crim.App.1977), for the proposition that although an experienced narcotics officer may identify marihuana, he may not testify that a powdered substance is heroin. The State agrees that no probative value should be given to this testimony as an expert, but argues that we should give value to his testimony as a nonexpert witness because he was testifying about his actual observations of the test results.
We believe that Malloy’s testimony about the performance and results of a field test is expert testimony. As such, in accordance with Curtis, he could not testify that the substance was cocaine. However, he could testify about the procedure he used in performing the field test, and about the physical results of the test. He stated:
The swabs are removed from the package, and they’re — you run alcohol over them, make sure that there is no foreign substance on them — the cotton. And then, you place the swab in the crack pipe, or whatever you’re testing. And if it turns blue, there’s presence of cocaine, which is what happened here.
*722Thus, although we give no probative value to Malloy’s conclusion that because the cotton turned blue, cocaine was present, we do assign value to the remainder of his testimony.
Further, unlike Curtis, the State had an expert chemist testify. “If this testimony (the ñeld officer's) was error, it was rendered harmless when a qualified and expert chemist later testified that the substance was cocaine.” Hicks v. State, 545 S.W.2d 805, 810 (Tex.Crim.App.1977). In support of his opinion that the combined substance from the three pipes was cocaine, the chemist spoke of the significance of the field test and the reliability of the field test results:
Q: How reliable is that test?
A: Well, it’s a presumptive test.
Q: What do you mean?
A: There can be false — there can be false positives, meaning that a substance other than cocaine can produce the color, in this case being blue; and the substance could turn out to be not cocaine, in many instances.
Q: What substances will cause that result?
A: Well, laundiy detergent; baking soda; some other drugs that are not controlled substances could also cause the blue color to be produced.
[[Image here]]
Q: Now, the officers used the C-Pak Kit to make the determination in the field, correct?
A: That’s my belief, yes.
Q: Of the metal pipes you received, have you ever seen any false readings?
A: No. Not on metal pipes.
Q: And glass pipes?
A: Maybe one or two out of many thousands. ...
Q: And it’s your testimony, that crack cocaine was positive in this?
A: In — yes. I found crack cocaine.
The chemist further testified that he performed two more tests, an ultraviolet spec7 trophotometry and an infrared spectrophotometry, both of which verified the presence of cocaine from all three tubes.
Appellant argues that because the chemist analyzed a combined substance from Pipes One, Two, and Three, the evidence only presents a plausible explanation of appellant’s guilt. We disagree. The record supports conflicting inferences. Malloy testified that he recovered Pipes Two and Three from the passenger’s side of the vehicle, and that upon field testing the pipes, the cotton swabs turned blue. The chemist testified that blue may indicate the presence of cocaine, and that he had never seen a false reading from a field test on a metal pipe. Together, the chemist’s and Malloy’s testimony combine to support a reasonable inference that at least some of the 7.06 milligrams of cocaine found in the three pipes came from at least one of the two pipes recovered from appellant’s side of the vehicle. In cross-examination of both Malloy and the chemist, appellant’s trial counsel raised a conflicting inference that all the cocaine could have come from Pipe One, rather than from Pipes Two or Three.
When we review a record which supports conflicting inferences, we must “presume — even if it does not affirmatively appear in the record — that the trier of fact resolved any such conflict in favor of the prosecution, and must defer to that resolution.” Matson v. State, 819 S.W.2d 839, 846 (Tex.Crim.App.1991) (citing Farris v. State, 819 S.W.2d 490, 495 (Tex.Crim.App.1990)). Accordingly, we find the evidence sufficient to support the jury finding that appellant intentionally and knowingly possessed cocaine.
We overrule appellant’s point of error, and affirm the trial court’s judgment.
Justice O’CONNOR dissenting.