Mr. Uelmen.

CROSS-EXAMINATION

BY mr. uelmen:

Mr. Matheson, the frequency tables that you rely on, you indicate, are based on the tests that are done in your own laboratory?

That's correct.

And you accumulate this data over a long period of time?

Yes.

Now, as a police laboratory, the tests that you conduct ordinarily are on people who have been arrested as criminal suspects or the victims of criminal activity?

That's true.

That's not a random sample that is representative of the population as a whole, is it?

No, it is not. It is, like you mentioned, victims and suspects in the city of Los Angeles.

And, in fact, ethnic or racial minorities would be substantially overrepresented in that group, wouldn't they?

Objection. Irrelevant.

Overruled.

Well, like I mentioned, we do not get racial information on most of the -- or all of the samples that we receive; but logic tells me, yes, that is probably the case.

By mr. uelmen:

For example, would you dispute that the African-American population in California is about 7 percent?

That's approximately what I have heard. I don't have any statistics with me.

Yet for arrest statistics and arrest data, they account for from 30 to 35 percent of arrests, don't they?

Same thing. I would have to accept your data on that, but I believe it is in that area.

all right. So minorities are much more likely to appear in the group of persons that you are testing for blood or bodily substances than they appear in the population as a whole?

That's correct.

So the frequency data that you have come up with may not be replicated if we were looking at the frequency in the population as a whole?

That's true.

Now, tell us a little bit about your laboratory. How many serologists do you have employed in your laboratory?

Well, first off, all of our professional staff are criminalists. They are not necessarily hired as serologists or toxicologists or narcotics chemists. Within the serology unit right now, I have six people working.

Now, do all six of these people work in the same laboratory?

Same physical location, yes.

In one room?

Yes.

all right. And all of your tests are performed in that one room?

Yes, that's correct.

Does your laboratory follow the f.b.i. guidelines with respect to controls for contamination?

I would have to see exactly what they are. We do have a number of controls regarding contamination in place.

Controls for contamination are pretty important in a laboratory doing the sensitive kind of testing that you do, are they not?

Definitely.

And occasionally there have been problems in laboratories doing this sort of testing with respect to what are called blooms of contamination?

I am not sure what you mean by "blooms of contamination."

That is not a term you have heard before?

Not in relation to the type of testing I have performed here, no.

Have you heard that with respect to D.N.A. testing?

Objection. Irrelevant.

Sustained.

BY MR. UELMEN:

Well, you do D.N.A. testing in your laboratory, don't you?

Objection. Irrelevant.

Sustained.

Your Honor, we would offer to show the CONTAMINATION with respect to D.N.A. testing could affect the accuracy of results with respect to other forms of testing of blood and bodily substances as well.

Mr. Uelmen, there has been no evidence presented -- in fact, the prosecutor was specifically asked whether they were going to be presenting D.N.A. evidence in this proceeding; and the answer was "no." And, therefore, I don't think that is relevant at this time.

I am not proposing to offer D.N.A. testing results. What I am proposing to explore is whether there has been a problem of contamination in the laboratory in which Mr. Matheson is employed.

Your Honor, the witness has already indicated that the type of contamination Mr. Uelmen is referring to does not occur in the testing he has conducted in this case.

As your question relates to D.N.A. contamination, the objection is sustained.

BY mr. uelmen:

Let me ask you, more broadly, have there been any problems of contamination in your laboratory.

Objection. Irrelevant.

Overruled.

To my knowledge, we have not had a problem with contamination within our laboratory.

by mr. uelmen:

Have there been documented examples of contamination in your laboratory?

Not to my knowledge.

And do you do a regular regimen of testing to see whether you have any problems of contamination in your laboratory?

Well, every one of our tests has built-in controls that would indicate whether or not there is contamination or any other sort of quality control problem.

all right. Do you regularly do proficiency testing of your results?

Yes, we do.

And do you have an error rate compiled for your laboratory?

Of the proficiency tests, both external and internal, that we have performed, there have not been any errors.

What kind of external testing do you have of your proficiency?

We subscribe to two different services. The one that has been around the longest, that we have used the most, is a company called CTS, or corroborative testing service.

And do they use blind testing?

No, they do not.

So you have never utilized any blind testing of your proficiency by external sources?

Well, I think I need to clarify how you are using the term "blind." We receive the items. We know that they are proficiency test samples. We do not know the results prior to our completion of the test and submission to them, but we do know they are proficiency test samples.

So there is no routine process whereby you test samples without knowing that they are part of a proficiency test?

That's correct.

Now, all of the tests that you have described today that you did on the three blood samples and the bloodstain are exclusionary tests; is that correct?

If you mean by that -- yes. We attempt to exclude somebody.

That's correct. There is nothing here that would individualize a stain to any one particular person.

So any attempt to analogize this to fingerprints or precise identification of a person would be inaccurate; is that correct?

That's correct.

Now, the tests that you performed produce some sort of product, or result, that other people can look at and interpret besides yourself?

In the case of the Electrophoresis test, yes. We photograph the results.

You photograph the plate on which the test is performed?

That's correct.

And did you do that in this case?

Yes, we did.

all right. But there is nothing preserved with respect to the abo typing?

No.

So no other person could interpret the results that you receive from that?

Well, there is my raw data; in other words, the readings that I had as I was performing the test itself. Beyond that, no. All that is left is some additional sample that could be tested by somebody else if need be.

OKAY. Now, normally the abo testing, for example, is going to be more accurate if you are testing whole blood samples, is it not?

I don't know if it would be more accurate. Once we supply an answer, the answer, in my opinion, is accurate. The whole blood samples tend to be easier to analyze than a stain.

all right. And would that also be true with respect to the other tests that you performed here, the group 1 and the pgm subtype?

No; not necessarily. With those tests, we take them, basically, the same point where we are drawing down a part of the liquid sample to create a stain; and then we test the stain.

Is there any relationship between the age of a bloodstain and how successful these tests are?

Yes, there is.

And what is that relationship?

Well, as a stain gets older, the likelihood of getting conclusive results becomes less and less. The factors we are looking at are biological in nature and, unless they are stored in the proper condition, will eventually rot and be untestable.

At what point, for example, would you be unable to successfully conduct a pgm subtype test?

Like I said, it varies depending on the conditions. I have not been able to detect pgm on a stain that was picked up out at a scene as short as a week later. In that particular case, the stain was subjected to direct sunlight on the pavement for in excess of 24 hours, I believe. I have also, on the converse side of that, successfully gotten pgm on stains that were, you know, six months old but had been collected fairly quickly and then stored frozen.

So the promptness with which the samples are collected and how they are stored can have a significant impact on these tests; is that correct?

That's correct.

Now, you were not able to get conclusive results with respect to item 49 on all of the tests that you conducted, were you?

That's correct.

all right. Could you explain which of the tests you conducted gave you an inconclusive result?

I am going to refer to my report, a copy of my report. In relation to item No. 49, I got no activity for the glo test, or the glo enzyme.

all right. The glo was part of the group 1 test that you did?

Yes, that's correct.

What would account for that, that you would not get any results on the glo on the group 1 test?

The most likely answer is that some degradation of the sample had occurred. The glo, or glo enzyme, degrades very rapidly; and we probably get less, or fewer, conclusive results from that one enzyme than from any of the others.

Does the fact that you get inconclusive results on the group 1 test in any way raise doubts as to the results you get on the pgm subtype test?

You mean as far as -- I am assuming at this point you would be talking about the pgm result that is derived during the course of the group 1 testing, and "inconclusive" merely means there is some question about that particular reading; and if it was a straight inconclusive, or no activity, then I would be concerned a little bit about the result that was obtained on the subtype test. We like to have a confirmation. In this particular case, item No. 49 indicated a 2+2-. I would expect the pgm to be a 2. They correlate there.

Now, the pgm test is the same test that you rely on for the glo enzyme, isn't it?

Well, it is All part of the same run on the same Electrophoresis plate. It is not necessarily the same test. It just so happens you can develop different portions of this same gel for different enzymes.

Now, you make one assumption in the test that you conduct on item 49; and that is that the substance you are testing emanated, or came, from one individual?

I don't assume that while doing the test, no. I run my analysis. I do assume that when I am doing the frequency calculations, that's correct; But as far as the rest of the testing goes, it doesn't matter whether it is a mixture or not.

Are you saying you would get the same results you are reporting here even if the stain that was being tested was a mixture that came from two different individuals?

Well, this particular stain gave the results that were obtained. If there is more than one individual present in that, it would still give the exact same results because that's what was obtained on the sample. If you are saying that this sample was mixed with the blood of another individual that had different genetic marker types than are exhibited here, then the results would have reflected that.

Let me give you a hypothetical. Assume that we had an assailant with type o blood who was injured in the course of the commission of a crime and on that assailant's hands was blood of the victim, which is type a. If those two bloods were mixed and the drop that you are testing was some mixture of those two blood types, what result would you get from your abo test?

As far as the example you gave, the results would be the same here. It would be indicative of a type a.

all right. So you wouldn't know that there was a type o assailant because you are assuming that this all came from one person?

Getting back to the other question, I am not assuming it did come from one person; however, the results that you obtain during the testing of a bloodstain in the abo blood typing system, if it is strictly from a type a person, you can have the presence of the a antigen or a combination of a and h, which ultimately indicate a type a. A type o has this h antigen in their system; so if you have a mixture of blood from an a person and o person, you would also see the a and h and, ultimatley, it would be indicative of a type a.

So you would get the same result?

That's true.

So if the scenario I presented involved an assailant with type o blood and a 2+2- pgm mixed with the blood of a victim who had a type a blood and a 1+ pgm subtype, would you get the same result you have here?

No. Given the scenario you just described, in the pgm blood type system, I would expect to see a 1+2+2-.

Is there any quantitative variation in that mixture in which you would not see the 1+ pgm subtype?

Sure. At some point, if the mixture of blood that was the type a, pgm subtype 1+ got down low enough so that the 1+ was no longer detectable, then theoretically that can happen.

What kind of mixture would that be? 20 to 1?

I have no idea.

No idea?

No.

So if there were a possibility of mixture, that would substantially affect the calculation of frequency of occurrence that you have presented here today, wouldn't it?

If we were -- if it was required to calculate a frequency, then we would also have to include in that frequency the population type o individuals.

all right. Well, what difference would that make if we included both o and a in the abo results and then multiplied it by the frequency you are using for pgm subtyping?

If you give me a moment, I will calculate that for you.

Thank you.

all right. So instead of 1 in 200, it would be 1 in 100?

That's correct.

Approximately. That's correct, yes.

Thank you. Nothing further.