Thank you.

Now, Dr. Cotton, please excuse my handwriting--printing.

First of all, your Honor, if I may, I just want an acknowledgment from the jury that they can at least see it from where they are.

1492?

I can see most of it except the "contamination" part.

I mean the line part.

All right. Proceed.

Now, Dr. Cotton, when we left off, you said that, for instance, if the stains at Bundy came from older blood drops, that would still be consistent with the results that you got and the numbers that you generated, correct?

That's right.

And you also said I believe that if in fact while the Bundy swatches were in the evidence processing room and they were cross-contaminated--they were degraded and cross-contaminated with the Rockingham swatches, that also would give you the same result that you have, correct?

If you had enough contamination, yes.

Even if the Bundy drops initially came from someone other than Mr. Simpson?

Yes.

And also, you--well, we didn't get to this before, but we've now done this during the break. If in fact on the morning--on--during the day of the 14th, when the Bundy swatches were in the same evidence processing room with Mr. Simpson's reference sample, fresh reference sample at a time when the reference samples of Mr. Goldman and Miss Nicole Brown Simpson were not there and they came in contact with one another, that too could produce the same results that you have on through DNA typing at Cellmark, correct?

Yes.

Excuse me. Objection. Assumes facts not in evidence.

Noted. Overruled.

I'm sorry. I didn't hear your answer.

All right. And again, you also said that obviously if there's a coincidental match, in other words, if somebody else has the same profile as Mr. Simpson, that too would be able to generate the results that you got in the laboratory, correct?

Of course.

And finally, if there was some other type of contamination at another time, another place, either unintentional or deliberate, that too could produce the same results that you have on your DNA testing, correct?

Depending on what the contamination was from, it may or may not.

Okay. Well, assuming that the type of contamination was the same type as described in item 2 here and item 3, but it occurred at a different time and different place and perhaps not so benign, somewhat malevolent--excuse me--nonetheless, would it too generate the same result that you got from the DNA profiling in this case?

Yes.

Okay. Now, you said that you don't know the probability of it being old drops, correct?

Well, objection. Assumes facts not in evidence.

Sustained. Rephrase the question.

All right. Assuming that they were old drops, you have no idea what the probability of that occurring is, do you?

Same objection.

She's already been asked and answered that question.

Would it be fair to say for all of these explanations, the only one that you can give a frequency estimate on is that of a coincidental match?

Of course.

Okay. Now--I'm going to return--

And I believe next in order would be 1144; is that correct?

1144.

Thank you.

Now, Dr. Cotton, as an expert, you are often called upon to interpret data; is that right?

Yes.

I'm sorry, your Honor. Sorry to interrupt. Are we finished with the exhibit?

No. I'm going to come right back to the exhibit with these next questions.

All right.

And sometimes, there may be more than one explanation for a certain piece of data; isn't that right?

You mean in terms of the data that we're looking at for any specific sample set?

Well, no. Just in terms of data as a scien--let me give you an example, a specific example. I believe you said on direct examination--

Can you put this up?

Now, putting up on the elmo People's exhibit 257-d, I believe you said on direct examination that there may be more than one explanation, for instance, why some bands are faint and others are darker?

Yes.

However, you said the most common explanation is that there is less--it's because there's less DNA in that lane; is that correct?

That's right.

And then I think you said also that there are other secondary explanations, but they are not as common; is that correct?

There are other secondary explanations that could contribute to a band either being darker or lighter than some other band.

But the most common explanation is because there's less DNA in the lane?

Yes.

Now, likewise, Dr. Cotton, in the various explanations that appear on Defendant's exhibit--which is right next to you at this point--1144, if any of those explanations occur with greater frequency than the 1 in 180 million, would those other explanations deserve greater attention by you?

Objection. Argumentative.

Sustained.

I can remove the exhibit. Since they'll be more drawing, I'm just going to turn the page.

I believe you said on direct examination that if you were to open up a cell, a human cell that contained chromosomes and you took out all the DNA out of that single cell, there would be approximately three billion base pairs; is that correct?

No. I said six.

Is that six billion base pairs or six billion bases?

Six billion base pairs.

Okay. And I believe you also said that approximately 99 percent of all of our DNA is identical; is that correct?

That's a rough estimation. In fact, it's probably somewhat greater than 99 percent.

And when you say that--let's say for argument sake, 99 percent is the same in all of us. You mean that the DNA that defines us as a species, as human beings as opposed to a lobster, for instance, is all the same?

Yes.

And so the DNA that decides that we have hands and not claws is going to be the same in all of us who are sitting in this room?

We hope so.

It's an eastern thing, your Honor. Now, you have Maine.

Well, I just said some lobsters.

All right. But it's that remaining one percent more or less where there is some variation within the human population, right?

Yes.

And that one percent, according to your estimate, would be approximately 60 million base pairs; is that right?

I think that's right, yes.

And your tests that you do at Cellmark don't compare the 60 million base pairs showing variation within the human population, do they?

Of course not.

When Cellmark does a forensic test, it examines how many different markers within the chromosomes?

Objection. Vague.

Overruled.

When one does an RFLP test.

We have at our lab the capability of looking at five. Some labs have the capability of looking at more, but that's what we're currently doing.

And for your PCR markers, you look at how many?

Six. Well, we now have the capability of looking at nine. But for this case, we looked at six, and so six would be the relevant answer.

And so would it be fair to say that there are millions and millions of base pairs showing variation that your tests don't examine?

Of course.

Now, when you say that a particular marker of DNA--that everyone is different for some of these markers, don't you mean that for a particular genetic marker, there is some degree of variation within the population?

Well, I didn't say that for any marker, everyone is different. But--and so your second statement is the accurate statement.

Okay. And would you also agree--I think you said it on direct--that for some of these markers, there's very little variation in fact in the population, whereas, for some of the RFLP markers, there is considerably more?

That's right.

For instance, for some of the individual markers that make up your poly-marker tests, one of them is called LDLR; is that right?

Yes.

And how many different versions are there of the LDLR marker?

In that--as that test is able to detect it, two.

All right. So would you agree that, for instance, there's more diversity in eye color within the population than there is for the LDLR mark that you use for forensic testing?

Well, it's probably not a great comparison, but in--the point is correct.

Well, would you agree also that for--that there's not a single RFLP that you actually utilize that would be unique to any one person? I guess that's redundant. Would be that one person would have and that no one else would have? Is that a fair statement, that's a single RFLP?

I don't think that there's--I don't think we actually know the answer to that question.

Well, for any of the RFLPs that were utilized in this case for which you profiled Mr. Simpson, is there any particular band that you have found that is totally unique in the human population?

As the bands are measured in this test, there are probably not ones that are unique in the population.

Okay. And I believe you also said that the first thing you do when conducting a DNA test and looking at the autorad is, you make a visual comparison; is that right?

It's not the first thing we do when we're conducting a test, but--

No. After looking at the autorad.

--the first thing we do when you have the film off is to make a visual assessment of what's on that film.

And correct me if I'm mistaken, Dr. Cotton, because I want to get this right, but if they are visually indistinguishable, do you then declare them to be a match?

No.

Do you declare them to be--to appear to be consistent with one another?

Well, when--at that point, we're not declaring them to be anything because at that point, then you're going to go ahead and do the computer imagining system assessment of band size and then you would make an assessment as whether you had a match or not. So--

And so would--I'm sorry. I didn't mean to interrupt.

So you're not writing anything down. You're not saying I have this opinion. You're just going forward.

All right. So--and you go forward to the computer because would you agree that it's necessary to have some objective and quantitative procedure for measuring the size of these bands?

The computer gives you a quantitative procedure and it--and it assists you in having an objective procedure, but it doesn't completely provide an objective procedure because the machine isn't smart enough to make the judgments that the human eye can make.

So it's a combination then of the person with the computer that gives you an objective and quantitative procedure for measuring the size of the bands?

Yes.

Now, you mentioned that given the limitations of the system, the imprecision of the system, you said that no band would be unique in your system; is that correct?

For a single genetic location.

Right.

I mean we don't really know. I'm just saying, based on what we do know and our experience, I wouldn't want to say that there was one that was unique. So--

Well, you mentioned, when you're talking about imprecision, that if bands don't necessarily line up exactly, in other words, one band could have more base pairs than another band, under your system and given its imprecision, you would still declare a match; is that correct?

If they're close enough in their measured number of base pairs, yes.

All right. Let me ask you, just to illustrate this point, assume this is the autorad and you have a band in lane 1, all right? And that band is--let's call it 10,000 base pairs. In other words, your computer has looked at that band and measured it and determined that it is a 10,000 base pair band.

Fine.

All right? Okay. Now, given your system's imprecision, how much bigger could a band be or smaller could a band be than 10,000 base pairs and you still declare it a match under your procedure?

If you want the precise answer to that question, I need to look in our standard operating procedure. But if you'll accept something that's reasonably close--

I will.

--the figure would be about four percent. There's a whole set of figures. So I'm trying to remember what the figure is in that size range, but four percent is going to be close to whatever the actual figure is.

Okay. Now, four percent of 10,000 is approximately what?

400?

Yes.

So if a band in lane 2 was let's say a little bit larger, 10,400 base pairs, under your system's imprecision, you would still declare a match between these two bands even though one has 400 more base pairs to it than the other; is that correct?

Yes. Given our--my example of four percent, yes.

Okay. Well, is that a correct approximation? If it's not, please give me another one.

I'm just telling you, it's close.

Okay.

I--I just don't remember the exact set of figures.

And, likewise, if the band was smaller than 10,000 and was only let's say 9,600 base pairs, under your system's imprecision, you would still declare a match between the 9,600 base pair band and the 10,000 base pair band; is that correct?

That's right.

Okay.

One moment.

Now, would you agree, Dr. Cotton, that within that range that I gave you there, 9,600 to 10,400 base pairs, there could be in fact, in reality that is, different and distinct alleles?

Of course.

Given the imprecision of your system, Dr. Cotton, that you've just described, if an item of evidence and a suspect have a band, a particular band which in fact differ by, in our example here, let's say 400 base pairs, then you would declare the suspect and the item to be a match, is that right, just for that band?

Yes.

And even if in fact the suspect's band represented a different allele than the evidentiary allele, you would still make that declaration of a match given your system's imprecision for that single band; is that correct?

Well, let me be clear here. In our laboratory, we don't ever report a match on--based on a single band.

Right.

But if you're asking me, do you have this size band and this size band and they're within the correct range, are they going to be called a match, the answer is yes.

Just so I can understand the system, and in fact in this particular case, when you actually did computer sizings--you did computer sizings of the bands in this case, didn't you?

Oh, yes.

And when you compared, for instance, a band in lane 52 and then you sized a band in Mr. Simpson's reference lane, the number of base pairs that your computer came up with was not the same number for both those lanes; isn't that correct?

They're not identical. That's correct.

Okay. Now--

One moment, your Honor.

Now, I take it, Dr. Cotton, that to arrive at a frequency of 1 in 170 million people, you did not test the blood of 170 million people?

That's obviously correct.

In fact, you didn't test the blood of even a million people; is that right?

Not in my lifetime.

Or even 100,000 people?

We tested, as I said before, for the database, the size ranges from about 150 to about 325. If you add all the people in the database up, all three racial groups, comes to around 5- to 600.

And it's taking a number--well, let me ask you. You gave a number though, for instance, for Mr.--for the profile, Mr. Simpson's profile, you said you expect to see it in 1 out of 170 million people in the African-American population; is that right?

Yes.

Okay. And how large is your African-American database from which you generated that number?

If you'll hang on just a minute, I'll tell you.

Certainly.

I'm ready.

Okay.

The--what I'll do is read you the sizes for the number of people that are in each probe one by one because they're not all the same.

Well, let me ask you this then. You mean, there's not--is there a single group of people who your laboratory has identified as--as African Americans on whom you ran all five of the probes that were utilized in this case?

No. There's a lot of overlap between who's, for example, been tested with MS1 and MS31 and so on. But the samples do not have--most of the samples do not have all five probes across each sample.

All right. Can you tell us how many of the samples you did run all five probes on in the African-American population?

Not without going back and doing a lot of figuring from the database. I don't know that off the top of my head.

All right. What I would simply ask is that between tonight perhaps when we break and tomorrow, maybe you can make some phone calls and there's someone back at Cellmark that can help you learn that so I can ask you tomorrow.

I'll try to get that information. I don't know--I don't know--I haven't been at Cellmark for a while. I don't know which of my lab staff are available to help me find things in the lab.

Thank you very much.

Take into account the time difference too.

Certainly.

Well, putting that aside and starting out with your premise that you can't really say how many people were subjected to all five probes, why don't you give me the range then for within the--your African-American database?

So would you agree that at a maximum, the greatest number of African Americans in your African-American database who have been subjected to all five probes would be that 140 something number you gave for YNH24?

146.

146 people?

Yes.

Okay. And so based on your analysis of 146 people of African-American ancestry, you arrived at a combined frequency in this case of 1 in 170 million people; is that correct?

No, it's not because some of the probes had quite a large number, more than 146. And since you're looking at the frequency of bands for each individual probe, the frequencies are not--the fact that you don't have each person across all five probes doesn't affect your ability to look at a frequency for a specific probe.

Fine, Dr. Cotton. So it's 200--what's the largest number--what's the largest number in the database for any of those five probes?

240.

So what you did was, you looked at the DNA profiles of 240 people, and from that, you extrapolated a number for the frequency of Mr. Simpson's DNA profile of 1 in 170 million people; is that correct?

That's correct.

And would you agree that there has been some controversy about the appropriateness of using databases that size for extrapolating these kinds of large numbers?

Well, excuse me. Objection. Hearsay.

Overruled.

Also irrelevant.

Overruled.

I don't think that the controversies have arisen around the size of the database and there is literature to support databases of this size for this purpose.

Well, let me go back--

One moment.

One moment, your Honor.

I just need to get the diagrams that Dr. Cotton made yesterday.

I think they're maybe with the court clerk.

Which diagrams? The hand-drawn diagrams?

The hand-drawn diagrams.

Now, to compile this African-American database, did you seek subjects--did you seek subjects from different countries in Africa?

No.

Did you seek subjects from different African-American communities scattered throughout the United States?

No, we did not.

In fact, your entire database--your entire African-American database comes from the red cross--I'm sorry comes from the red cross in Detroit, Michigan; isn't that right?

Yes, it does.

And would you agree that to go from a database of 250 African Americans in Detroit to a number of 1 in 170 million to one in 1.2 billion, you had to make certain assumptions about the independence of these different genetic characteristics?

There are a whole series of assumptions built into that calculation, and one of them is independence, but there are also tests for independence in these databases which have been done and they have been shown based on these tests that the markers are independent.

Now, Dr. Cotton, would you agree, however, that they are based on a number of assumptions, one of which you said is that these different genetic markers are inherited independently of one another?

Yes. But what I'm saying is that there is data to back up that assumption and, therefore, it is not really an assumption at this point.

Would you agree, Dr. Cotton, that there is--or I'm sorry--would you agree, Dr. Cotton, that there has been substantial controversy in the scientific community about that various assumption that you've just described?

Same objection. Also hearsay.

Sustained. Why don't you rephrase the question. It's vague.

You were talk--

One moment.

Referring specifically to the assumption of independence, would you agree that with respect to that assumption and whether or not there is adequate data to demonstrate that it's an appropriate assumption, that that itself has been the subject of significant controversy in the scientific community?

Objection. Assumes facts not in evidence.

Sustained.

I'm asking her if it is.

I understand. Sustained. Why don't you rephrase the question, counsel.

All right. Well, Dr. Cotton, had there been controversy in the past about that very point?

Yes.

In fact, Dr. Cotton, that book that you had in your purse, what--or your briefcase, could--what--could you take it out again, please, the NRC report?

The NRC report?

Yes.

The NRC report says that there's controversy around that various assumption--

Excuse me. Objection. Hearsay, your Honor.

Sustained. Sustained. The jury is to disregard the implication of the question.

Dr. Cotton, isn't there an entire chapter--

Excuse me. Objection, your Honor.

Sustained. Sustained. It's hearsay.

Dr. Cotton, you said that--

What's the nature of the controversy, doctor?

There have been a few people who are experts in population genetics who have not agreed with the way statistical calculations are made for forensic casework as given in the RFLP example. There are also a large number of people who agree that the statistical calculations are appropriate. So depending on who you phoned up, you might find an opinion one way or the other.

So are you saying that you disagree with what the NRC says--

Excuse me. Objection. Hearsay, your Honor.

Sustained. Sustained.

Well, you said there are a few people, Dr. Cotton, who take that position. Isn't it a fact, Dr. Cotton, that as recently as November of 1994, more than two dozen --

Excuse me. Objection, your Honor.

Sustained. Sustained.

May we approach?

I'm asking if she's aware.

No. You're calling--the question and answer calls for a hearsay conclusion as to the content of something, counsel. That's the problem. Sustained.

When you said, Dr. Cotton, a moment ago that there's a few scientists who are critical of that assumption, how many did you mean by a few?

Again, objection. I think this is also hearsay also, your Honor.

Overruled.

Based on the number of published articles that state an opinion one way or another, I would say there might be--this is sort of a little hard to answer, but maybe something around five or six people who have been vocal in their opposition to these calculations.

All right. Dr. Cotton, let me ask you this.

Excuse me. Could the witness finish her answer, please?

The question--

Yes.

I think that was responsive. And beyond that would be--

Well, she was still speaking when you jumped in there, counsel.

Sorry, your Honor.

And there's an objection. When the objection is made, you need to allow me to rule on it. Did you finish your answer, doctor?

Yeah, I think so.

All right. Next question.

Do you know who Dr. Ronald Acton is?

Yes, I know Dr. Acton.

Has Dr. Acton written on the subject of population frequencies?

I'm aware of one paper that he's written.

Do you know who Dr. David balding is?

Vaguely.

Do you know who Professor Joel Cohen is?

Objection. Relevance at this point, your Honor.

Overruled.

I've read one paper by Dr. Cohen.

You've heard of him?

Well, through seeing his name on--as the author of a paper.

Do you know who Professor and Dr. Jerry Coyne is?

I've heard of him. I haven't written--I mean, haven't read anything that he's writ--I'm not aware that he's written anything on this topic.

Is he a population geneticist?

I don't know. I don't know. I'm sorry.

Fine. Are you aware--are you aware of any testimony that he's given, any cases without saying what it is?

Objection, your Honor. Irrelevant and hearsay.

Sustained.

Do you know who Dr. Peter Donnelly is?

No.

Do you know who Dr. Seymour Geiser is?

Yes.

Is he a Professor at the University of Minnesota?

Yes.

And Dr. Joel Cohen is a Professor at Rockefeller University?

Objection as to relevance, your Honor.

Overruled. Do you know that?

I don't know.

All right. Proceed.

Do you know who Dr. Paul Hagerman is?

Yes.

He is a Professor at the University of Colorado?

Yes.

You know who Dr. Daniel Krane is?

Yes. I've read one of his papers. When I say yes, I don't know these people personally, but I may have either heard of them through court issues or read something that they wrote.

Do you know who Dr. Martin Kreitman is?

No.

Do you know who Dr. Charles Langley is?

No.

Do you know who Dr.--Professor Richard Lewontin is?

Yes.

He's a Professor at Harvard University?

I understand that's where he is.

He's a population geneticist?

Yes.

Do you know who Gary Littman is?

No.

Do you know who Michael lynch is?

No.

By the way, I didn't say Dr. Before each name, but just so the record is clear, each one of these is a doctor.

Well, excuse me, your Honor. Counsel is testifying now.

He is. Proceed. The jury is to disregard that as--

Do you know who Dr. Teri Markow is?

No.

Do you know who Dr. Lawrence Mueller is?

Yes.

Is he a Professor at the University of California at Irvine?

As far as I know, yes.

In what? In what?

In the Department of Ecology and Evolutionary Biology?

Well, excuse me, your Honor. Is counsel testifying again?

No. I'm putting it as a question.

Question.

Do you know if he is a Professor in the Department of Ecology and Evolutionary Biology?

I think so, yes.

Do you know who Dr. Richard Nichols is?

No.

Do you know who Dr. Stephen Peiper is?

No.

Do you know who Dr. Rollin Richmond is?

No.

Do you know who Dr. William shields is?

Your Honor, I'm going to object again as to relevance.

Overruled.

Yes.

Is Dr. William shields a Professor of Population Genetics and Genetics at the State University of New York at Syracuse?

Yes, he is.

And do you know who Dr. Terry speed is?

No.

Do you know who Dr. Charles Taylor is?

Yes.

Is Dr. Charles Taylor a Professor of the Department of Biology of the University of California in Los Angeles?

He was last time I was aware of him.

And have you heard of Dr. Elizabeth Thompson?

Yes.

And she is the Chairman of the Department of Statistics at the University of Washington in Seattle?

I know she's at Seattle. I have no idea what her position is there.

And do you know who Dr. Michael Turelli is?

No.

Do you know who Dr. Donald--

One moment.

Excuse me. Do you know who Dr. Donald Ylvisaker is?

No.

And do you know who Dr. Sandy Zabell is?

No.

And finally, do you know who Dr. Marcus Feldman is?

The name sounds familiar, but again--

Would it refresh your recollection if you knew that he was a Professor in the Department of Biology at Stanford University in California?

Excuse me. Assumes facts not in evidence.

Sustained.

I'm sorry. It doesn't help.

Now, hypothetically, Dr. Cotton, if all the people on that list--

Excuse me, your Honor. Objection.

No, no, no, no, no, no.

Dr. Cotton, at any time in the last five months, have you seen a letter--

Excuse me.

I'm not saying the contents of it, your Honor.

Objection. Irrelevant. 721.

Sustained.

Dr. Cotton, at least as to the people whose names you recognized, do you agree that they have articulated that there was--

Objection, your Honor. Hearsay. Could we approach sidebar?

No. Sustained.

I would ask that this be marked next in order to show it to the witness.

What?

A letter.

All right. Next in order.

No. Just hand it--

Could I see the exhibit, your Honor?

Sure.

All right. For the record, is this a letter in science magazine?

No.

What is it?

May we approach, please, your Honor?

No. What is it? Just tell me what it is.

It is--I'll tell--

Mr. Neufeld.

May I have my exhibit?

I would like to look at it first.

Okay.

Your Honor, this is apparently a letter--

Your Honor, now you want a speaking--

Counsel, what is it?

It is a letter written to nature signed by 25 scientists.

Thank you. Thank you.

Well, your Honor, that was rejected and never published.

Your Honor, it was rejected because--

Wait.

That was--

Both of you.