All right. Thank you, ladies and gentlemen. Please be seated. Let the record reflect we have now been rejoined by all the members of our jury panel. And Miss Renee Montgomery is on the witness stand undergoing cross-examination by Mr. Blasier. Mr. Blasier, you may continue with your cross-examination.
Miss Montgomery, I just want to clarify that the item--the band-like thing down here that we have been talking about corresponds to the lane at the top for the substrate control, does it not?
Correct. I'm sorry. Actually you should also include that extraction blank in the second--or in the lane next to the ladder because since it is right in between those two lanes you need to address both those lanes.
I want to ask you a couple of questions about the D1S80 system in general that you talked about on direct. You indicated that it was a discreet allele system, correct?
What did you state with respect to the ladder encompassing all the possible alleles?
Okay. The ladder is a composite ladder and that means that from the 14 allele and then the 15 through the 41 allele are present on that. There are alleles that are greater than the 41 and there are also--there is possibility that a 15--a 15 allele has been seen, but it is very rare, and so it is not added into the ladder.
So there are alleles in this system that don't show up on this ladder, correct, because the ladder is too small?
There are some very rare alleles, like I said, greater than 41 and then also the 15 allele that are not on this ladder.
And those are alleles that could be present in someone's DNA that would not show up in this system, correct?
No. That is not correct. They are--they would not have a corresponding ladder allele to call them. If there were something greater than the top 41 allele, and you would see a band up there and I have seen bands greater than 41 and this is very rare and--so anything that is greater than 41 is called greater than 41 and it can't be lined up with any allele ladder. And it happens so infrequently that they have--the company doesn't feel it is necessary to have ladder alleles up in that region, which I agree with.
But the point you indicated is that alleles don't run off the gel. There are alleles in this system that do run off the ladder in the sense that you can't--you can't measure them, correct, with the ladder?
Now, when you say a discreet allele system is a good analogy there, comparing a digital watch with an analog watch that goes around, a discrete allele system is like a digital watch, meaning that you are either at one second, two seconds, three seconds, you can't be in between? Is that what you mean by discreet allele?
And a clock, for instance, is more like an RFLP system where you can have alleles spread out from the top to the bottom of the gel. They are not going to be in discreet positions?
Or continuous. Or another analogy would be shoe size. When you go in to buy shoes, they have eight and a half, nine, nine and a half, but you might be in between that eight and a half and nine, so for RFLP that is how RFLP is addressed, whereas with D1S80 each person would fit perfectly into that.
And that would be because that everybody who is a 25 has the same sequence of DNA at that location as anybody else that is at 25?
Well, they have the same number of repeats, so an individual that is a 25 would have 25 repeats or an individual that is a 25 would have 25 continuous repeats of that particular sequence. There can be subtle variation, but--and an individual that would be an 18 would have 18 continuous repeats of that region.
Well, when you say there can be subtle variations, what you are saying is that there can be people who show up as a 25, for instance, that have a different sequence than someone else that shows up as a 25, correct?
Oh, yes, that is possible. What we are looking at is length differences; we are not looking at sequence differences with this system.
So two people that come up as a 25 would match in your system, but if they had a different sequence they are different people, aren't they?
So your system does not have the ability, even though you call it a discreet allele system, to identify precisely the sequence in a particular band?
And unlike PCR and DQ-Alpha, which does identify specific sequences of DNA, correct?
So this in a sense is--this aspect of the system is closer to the RFLP system where you are just--you are estimating sequences in a sense, correct?
Well, we are not estimating sequences. We are not not looking at the sequence. We are looking at length differences between individuals, how many repeats they have. It is a 16-base pair repeat, so an individual that is an 18 has 18 of these 16 base pair repeats and they would have a repeat sequence that is, say, this long, whereas an individual that is a 25 would have repeats sequences that are even longer. The same repeat sequence, the same number of repeats, but just a different length.
The point is, different alleles would come up as the same, 25, 24, whatever and they would be different alleles, wouldn't they?
If there is a sequence difference between two individuals, they can run to the same position, but it is the length difference we are looking at, not the sequence difference.
Now, there is also reports in the literature, B bands in the D1S80 system not lining up correctly with the ladder, correct?
And there have been reports in the literature of bands going to a place on the ladder that is different from what the true sequence is, correct?
Repeating your question to myself. There is--if there is a sequence difference, it is possible that there could be just a very subtle variation, but it will still line up with the ladder, the composite ladder allele that is adjacent to it, provided in the Roche kit.
Well, the report--the literature indicates that there are situations where it is so far out of line with the ladder that it can be confused with the band above it?
Are you aware of the literature that talks about the mobility of bands being such that you can mistype which rung of the ladder you are on because the band size is somewhere in the middle?
Yes. That was early on during the development of this system, and what happened is some of the laboratories were trying to use their conventional serology--
I think she answered. The first part of her answer--I'm satisfied with the first part.
Now, this system also has a characteristic that if samples are degraded you can lose bands, correct?
You could lose your entire--you could get no results with degraded samples, yes.
You can lose--well, let me rephrase it. The ladder that is in the D1S80 system comprises fragment lengths that go from about, what, 300 base pairs up to a couple thousand maybe?
And when a sample degrades it generally starts to break down, the larger fragments get cut--get broken up first, there still may be shorter fragments while the degradation process is going on, correct?
Well, it is a random act that is occurring, the degradation of DNA, and that is just cutting at various locations on the geno.
So your understanding that when degradation occurs and DNA fragments are cut up, they are cut up randomly in the sense that--well, let me rephrase that. When you start chopping up the D.A.--the DNA you lose--
As it degrades further the bands that are longer, that represent longer sequences, you no longer can see, correct?
So you can lose the upper bands of a genotype? For instance, if somebody is an 18, 31 and the sample is degrading, you could lose the 31 and the person would appear to be an 18, 18, correct?
Well, theoretically that's correct, but from all the method--all the research that has been done and all the studies that have been done, what has been demonstrated is when you get to the level of degradation you typically don't see any result, and if you do see result, you will see something just so faint that you wouldn't be confident even calling that a genotype.
So are you saying that the kit has been developed to the extent that you would not see this particular phenomenon of the upper band disappearing?
Well, it is not the kit that has been developed to do that. It is just studies that have been done within laboratories to demonstrate whether this possibly can occur, and these are environmental abuse studies that have been done.
And that talks about one of the problems of this system being--one of the interpretation problems, that you may lose larger bands with a degraded sample, does it not?
Yes. This was their amplification conditions are used in our laboratory for our analysis.
Would you agree that that says that you may lose larger bands if a sample is degraded?
What it states is that when you have a single-banded pattern that your results should be interpreted with caution. The--the company has also done some follow-up research in which they've published a paper in May of--
And wouldn't you agree that the kit insert also tells you that some alleles don't align up with the ladder, referring to 8.3.2 right below that?
And does that indicate that there are a couple of reasons why that might happen as well?
And that gives several reasons why you might have bands that don't line up the way they are supposed to with the ladder, correct?
Now, the kit also is warranted for amplification of 2.5 nanograms of DNA or more, correct?
Yes. They state in here that they guarantee results with 2.5 nanograms of DNA.
And many of the tests that you ran in this case were on amounts far less than that, correct?
Now, the D1S80 system is less informative than the RFLP system because it has fewer alleles? Would you agree with that?
Well, we are looking at one region on the D1S80 system whereas with RFLP we are looking at various regions, as Dr. Cotton and Mr. Sims discussed.
One place in the DNA? With RFLP, with a single probe, you are also looking at one location in the DNA, correct?
And with RFLP, however, you can have many different alleles, 60, 70, 80 different alleles for a given probe, can you not?
So there is less variation among people in the D1S80 system than in the RFLP system because there is fewer possible alleles that you could find in people, correct?
Yes. I believe approximately thirty percent will have, within their genotype, a 24 or an 18 allele.
And that makes this system in this particular case where both Mr. Simpson and Mr. Goldman have 24 alleles, makes it less informative than if they had more rare alleles, correct?
Well, with Mr. Goldman, he is a 24 homozygote. With Mr. Simpson you need to look at the other allele that goes along with that 24 allele.
My question is when you have a 24 allele it is less--it is a less informative system than if you have a more rare allele, correct, because you can't narrow it down as far in terms of the potential contributors?
Correct. If you had two rare alleles, then it would be more informative, right.
It would be better. And the 18 allele, Nicole Brown Simpson is an 18, 18 and that is always very common allele, is it not?
Correct. By "Very common," as I just stated a few sentences back, that between the 18 and the 24 alleles, you will see this in one--that allele in approximately sixty percent of the population, either the 18 or the 24, but then you also look at the other allele that goes with the individual.
In fact, well if we do that with 18, 18, that is actually a fairly common genotype, is it not, 6 percent of the population has it, Caucasian population?
So of a hundred people you would expect there to be six people that would have that combination, Caucasians?
And that--because that allele is involved, that is much less informative than if you had a rare genotype?
Yes. If you had a rarer one, then you would have less occurrence in the population.
And the fact that Mr. Goldman shares an allele with Mr. Simpson makes it even less informative because there may be some difficulty in assessing who is responsible for the 24 allele? Would you agree with that?
Because there is an allele in common, that makes it--that makes it more difficult for you to separate components of a mixture because there is one allele that two people have in common? Would you agree with that?
Now, do you and Mr. Sims and the other people in your lab use the same standards to decide whether something is a hint versus a trace versus a C minus versus a C?
Yes. There is no standard technology as to hint, trace, for D1S80. So what happens is when someone is second reading the gel, they call it, and if they call it something different than the first analyst, then you document it in your notes. If they aren't calling it a hint, then you write what they are actually calling it.
Would you agree that that has a certain subjective element to it? One examiner might call something a hint and another person might call the same thing a trace?
It is possible. When you get down to those lower levels, umm, a hint would be--a hint uniformly throughout the laboratory by a DNA analyst--case work analyst is something that--there is a hint there. That is an indication that something is there, but you can't be confident that it is actually a band. A trace then uniformly through the lab is there is a band there. It is very, very weak band.
Now, you--on page 91 of your notes, that is a run where you were looking at what types of samples?
Now, would you agree that on--in your lane no. 12 on your work notes you described that as a 24, 24 with a hint of an 18?
I ask you to refer to your report at page--I'm sorry, April 6, 1995, page 8 of 11.
Well, a--there is a difference between actually writing your observations on your work sheet and doing your interpretation. Your work sheet is what you actually observed and then you--one needs to think about how they are going to word these reports, and with--I could have written in the report a hypothetical of an 18 allele. But a possible trace, the words sound more effective than a hint of an 18 allele, so the three of us that wrote this report, Mr. Sims, Mr. Myers and myself, discussed it and tried to come up with the correct wording that would adequately describe what we saw. And what we came up with was possible trace, and that is not saying that there is a trace of an 18. That is saying there is possibly a band there but we are not confident in calling it.
When you say you want to do it more effectively, in other words, to put forward the Prosecution's case?
No, not at all. To reflect what the evidence--to reflect what the evidence is. It doesn't matter if it helps the Prosecution or the Defense; we are just reporting the science.
What changed between your reading of the actual film and when you wrote the report to cause you to change the way that was described?
Did something change in terms of your analysis of this particular test result? Did you read it again? Did something change?
No. To write a hint in a report, it just doesn't--what does a hint mean in a report? Umm, in the laboratory I know what a hint means and I am able to hopefully tell you what a hint means. But in a report, for someone just to pick this report up and be able to read it, whether it is the Prosecution or the Defense, we need to convey what we are trying to get across to them and by saying a possible trace, that conveys that we are not confident that a band exists there, and to just say a hint of an 18, to me it doesn't state it a effectively as saying a possible trace.
Now, unlike RFLP technology, you don't use any kind of computer aids to determine whether there is a band there and whether it is a hint or a trace or possible trace, do you?
No. That is one of the great things about this system. We get to use the basics.
You just look at it and you say looks like a trace to me or looks like a hint to this guy, right?
And every analyst that looks at these things always comes up with the same terminology, hint, trace, possible trace?
Well, I can't--outside of our laboratory, I'm not quite sure if people would put the same words to these. It is--in our laboratory it is pretty well-defined, and as you could see on this particular one, three different individuals read it and they all came up with the same word, hint, at 18.
Have you ever done any studies where you have looked at different quantities of DNA in a mixture, for instance, to see if--to see what happens when you start reducing the quantity to determine at what point a band becomes an artifact?
Yes, I have done studies to demonstrate at which level or at what level you start losing a minor component.
And have you published any opinion studies that show the differences in intensity of the bands between a hint and a trace and a C minus?
It is made in our laboratory and it is just a scale, a progression, as I described earlier.
Now, as a sample degrades--have you done any testing to determine what happens when a sample is degraded in terms of how it shows up, whether it shows up as a band or an artifact?
Yes. I did some temperature studies in the laboratory to see what happened when various samples were subjected to various temperatures for differing amounts of time.
It is not like a light switch, is it, where you--it is either there or it isn't there? There are gradiations of band intensities, are there not?
You must use your subjective judgment to decide whether something is really a band as opposed to an artifact, correct?
Well, when one gets to the lower level where you are not sure if it is a band, then yes, that's correct.
Now, when you read these things, did you know what genotypes the two victims and Mr. Simpson had?
All the evidence samples that you read, when you did that, did you know the genotypes of Mr. Simpson and the two victims?
So you knew what results you expected to get, at least as to some of the contributors, correct?
You didn't have any kind of expectation in your mind that perhaps these evidence samples would produce these bands?
Well, no, because that is part of testing the evidence, is to show who can be excluded and who can't be excluded.
You know, I believe I have only heard it in the context of--let me back up. I--examiner bias I think would be an individual having some bias on their examination.
Have you ever heard it used in the context of one who knows the results they are supposed to get and subconsciously getting the results that they think they are supposed to do, not through any trickery or lying or anything, but you see what you think you are going to see?
KEY QUOTESomeone might interpret one thing as an artifact or as a faint band, depending on what they expect to see?
Now, the--you indicated that the range of size of the fragments in the D1S80 system was about 400 to 800 base pairs?
And the size of the fragments in the DQ-Alpha system are less than 400 base pairs, correct; it is about 240?
So would you agree that in a degrading sample you would expect the D1S80 bands to disappear the DQ-Alpha allele from a particular piece of DNA?
So that if you had a mixed sample, let's say in the DQ-Alpha, you see evidence of a mixture, but you don't see any extra alleles in the D1S80 system, that might be explained by the fact that the sample is degraded?
Yes. And one needs to also look at their yield gel to determine if degradation was occurring on that, if it is possible.
You are aware that the Bundy drops in this case were severely degraded or are you aware of that?
And are you aware of the DQ-Alpha testing results on the steering wheel sample, that is item 29?
And you are aware that there was a 4 allele which is inconsistent with a genotype of any of the three people in this case?
And the reason it was inconsistent is because there was no 1.3 to go along with that which would have been consistent with Ronald Goldman?
Right. There was a very low level amplification--or a low level amplification.
Because I remember discussing that issue with Gary Sims at the time that we were doing the analysis on those samples.
It is on Gary Sims page 40. There is another advantage to numbering pages, I guess.
No. It was a matter of there not being enough to test due to the fact that this sample was first used for DQ-Alpha, and after using a portion of it for the DQ-Alpha typing there wasn't enough sample left for D1S80 typing.
KEY QUOTEMiss Montgomery, could you look at the monitor and tell me if that appears to be a photograph of your run no. 297?
And the interpretation of this sample for samples G5 and G6 were an 18 allele stronger than a 24 allele?
And you agree that that is an accurate picture of that film for that particular run?
Yes, it is, but once--once again, it is difficult to see the weaker components of this and you will actually have to look at the gels yourself.
All right. Let me zoom in on G5 and 6. Would you agree that you cannot see any 24 allele at least on this picture?
Well, this is all done with your computer. I mean, this isn't the actual sample and so I don't know--
Would you look at your original sample and tell me if you agree that it is extremely faint?
But you can see it in here and when you do see a copy of it later on you will be able to see that it is a trace amount of a 24 allele.
Well, a hint, as I stated, is a--a hint is a possible. It is not a well-defined band. These trace, these are well-defined bands, they are just very faint, and as far as why you can't see it on your monitor, I can't explain that.
Yes, but I would prefer that it not be, umm, put into evidence. Mr. Harmon has a copy of the blue copies.
Now, the area of interest--now, the area of interest that we are looking at are the two lanes over toward the right between the two ladders, correct?
And you are saying that there is something more than a hint, which is a band at 24?
And I would like to have that shown to the jury, please, if we could mark that printout.
Actually, your Honor, let's hold off on that and show another picture and we can do a couple of them at the same time, if that is okay.
Now, Miss Montgomery, the picture that we have put up now, 1177, contains the samples from Nicole Brown Simpson's fingernail scrapings, does it not?
I have an objection, your Honor. May we approach on this? It relates to what Miss Montgomery said a moment ago.
the words sound more effective than a hint of an 18 allele, so the three of us that wrote this report, Mr. Sims, Mr. Myers and myself, discussed it and tried to come up with the correct wording that would adequately describe what we saw.
Have you ever heard it used in the context of one who knows the results they are supposed to get and subconsciously getting the results that they think they are supposed to do, not through any trickery or lying or anything, but you see what you think you are going to see?
No. It was a matter of there not being enough to test due to the fact that this sample was first used for DQ-Alpha, and after using a portion of it for the DQ-Alpha typing there wasn't enough sample left for D1S80 typing.
No, that is one of the great things about this system. We get to use the basics.