https://www.blogger.com/comment.g?blogID=6592607595936297457&postID=2951234322664707881&page=1&isPopup=true
http://anthraxvaccine.blogspot.com/2009/01/critique-of-chemical-signature.html
A possible solution to the Bruce Ivins FBI anthrax mystery is outlined in a comment at Meryl Nass produced below. This arose from an extended discussion of many but including Ed Lake and Attorney Barry Kissin, Ross Getman, some anonymous experts, and others. Perhaps Vahid Majidi and Director Mueller of the FBI could consider this.
Article by Paul de Armond linked to by Ed Lake :
http://www.albionmonitor.com/0208a/anthrax.html
Ed, thanks for the link and discussion. Of course, if you take everything in the article linked to as valid, it would tend away from Ivins in my view. But I wouldn’t hold anyone to such a standard.
http://en.wikipedia.org/wiki/Endospore#Formation_and_destruction
“
Reactivation of the endospore occurs when conditions are more favourable and involves activation, germination, and outgrowth. Even if an endospore is located in plentiful nutrients, it may fail to germinate unless activation has taken place. This may be triggered by heating the endospore. Germination involves the dormant endospore starting metabolic activity and thus breaking hibernation. It is commonly characterised by rupture or absorption of the spore coat, swelling of the endospore, an increase in metabolic activity, and loss of resistance to environmental stress. Outgrowth follows germination and involves the core of the endospore manufacturing new chemical components and exiting the old spore coat to develop into a fully functional vegetative bacterial cell, which can divide to produce more cells.
“
One article I saw in a search talked about 3 to 7 hours for one stage of this or maybe all 3. But some said it can be up to 60 days or never. The article I linked to above on runs found some dud runs. Even your linked article said it looked like multiple runs some duds were tried.
“
When a bacterium detects environmental conditions are becoming unfavourable it may start the process of sporulation, which takes about eight hours.
“
If the idea is to start Friday night and have spores ready Sunday morning, he is going to take 8 hours or more to activate to ready for exponential growth. But then its almost time to stop and go to work to put them to sleep, sporulation.
Ivins needed to be done Sunday morning to prepare the envelopes and clean everything up and get out.
Ivins knew the above biology already. He knew that starting with spores Friday night and ending with spores Sunday morning meant the time to grow would be zero. Moreover, he knew from long experience the time to activate was highly random and the initial growth highly random. The dud runs in the article I link to above show that.
The evidence from the article linked (by Ed) is that 10 percent of some of the first letters were spores and the rest growth material. Ivins knew that is what would happen. Ivins knew his schedule, and he knew spores to growing dividing cells back to spores by Sunday morning was a fool’s errand. So why did he do it? Dr. Popov in his discussion indicates it takes days to do anything meaningful including prep time. (There is also thorough cleaning to avoid detection here.)
Alternative scenario.
Someone someplace else as part of a conspiracy started growing spores the week after 9/11 on orders from someone. They were told to grow and prepare them.
Then the order came in to take what they had and mail them right away. That is, it was unexpected to stop the growth and mail what they had. That means it wasn’t Ivins. Ivins would have known he had to be back to spores Sunday morning.
So someone else in place X was told to prepare a lot of anthrax by growth. But then within a couple of days of that order they were told to stop and do sporulation and mail out what they had. This means it wasn’t Ivins acting alone.
Moreover, the place was not Ivins lab, it was someplace they expected to be able to run a large fermenter or shaker incubator for 7 to 14 days like in the paper I linked to. Run to completion and test for completion like the paper did. But then they were surprised and had to sporulate what they had and mail it. That means it wasn’t Ivins.
Its a classic detective story plot. (Too bad the FBI can’t figure these out isn’t it? But that is for the amateur detective?) We just needed to activate the little gray cells together mon ami. N’est-ce pas?
==
Searches that are interesting are
bacillus sporulation
and activation, germination, out-growth, anthracis, subtilis, etc.
==Prior posts
Old Atlantic Lighthouse said…
I would like to bring up the government’s own tests on how long it takes to incubate that they released in 2004 when they wanted to prove Ivins and others like him couldn’t do it in a prosecution of Hatfill.
http://oai.dtic.mil/oai/oai?&verb=getRecord&metadataPrefix=html&identifier=ADA426293
http://www.dtic.mil/cgi-bin/GetTRDoc?AD=ADA426293&Location=U2&doc=GetTRDoc.pdf
They used a
New Brunswick c25 floor shaker incubator
See page 12 of pdf, 2.9.2.
This weighs 400 pounds.
Note they were doing 1 liter flasks. Those produce in the best runs somewhat under a gram. To produce 5 to 10 grams for the first letters would require 5 to 10 liters.
http://www.artisan-scientific.com/59721.htm
http://www.artisan-scientific.com/info/New_Brunswick_C25_Datasheet.pdf
See page 2 for photo. The unit is bigger than a man and weighs 400 pounds.
There is a smaller unit
http://www.artisan-scientific.com/57635.htm
It weights 230 pounds.
http://www.artisan-scientific.com/info/Innova_4230_Datasheet.pdf
The paper above ran at 30 deg C using the incubator shaker. So if one didn’t use something that good it would take longer.
They started with 1ml of bacillus per liter of CD. So that is a factor of 1000, which is roughly 2 to the 10 power. So if you start with more you cut down the growth time needed. Even so, there doesn’t appear enough time.
Look at Table 6 on page 20 of the pdf.
They checked the runs every 24 hours to see how far along they were. The best run was 3 days to produce .5238 grams for the starting liter of CD. So to get 5 grams you need 10 liters of CD. Note even if you start with more than 1 ml of bacillus you still need the 10 liters of CD to get the 5 grams.
If we take 10 generations as 3 days, we get approximately 3 generations in 1 day, i.e. a factor of 8. Divide the 5 grams by 8 and we get 5/8 grams he needed to start with to get done in 24 hours. If he had a liter flask and the ratios are the same, then he needed to use 5/8 of his reference flask, if it was full to 1 liter.
Note, he could have put some back at the end.
If things didn’t go so well as this perfect run, it can’t work.
He also has to do centrifuging, rinsing, re-centrifuge, and eventually lyophilize.
Every piece of equipment he used to make it go faster had to be cleaned and put back where it belongs.
To do it all in one run and get the best possible growth times he had to have the best equipment.
Note if he had the 400 pound floor incubator, then if anthrax got on the bottom of it he couldn’t clean it. Even the 230 pound unit, he couldn’t clean.
Did the BSL3 have these big units?
As you consider all the other steps needed, including cleanup, intermediate steps, getting the CD ready, lyophilizing, putting it in letters, covering his tracks, etc. it becomes hard to believe he did it from Friday night to Sunday the first weekend.
Table 6 in the paper records total dud runs. Some runs take 7 days. For those runs even if he started with all of the RMR 1029 flask and ran 1 day, he wouldn’t get 5 grams.
Because at 7 days, you are doubling at 7/10 of a day. So if you double twice your at most 1 gram you get 4 grams. You then have to put 1 gram back, so you have 3 grams.
The paper above is sort of difficult to read and get everything out you need at one sitting. Its important to go over it more than once. That’s the benchmark.
To run 10 liters, Ivins needed a big C25 style. There are other fermenter units not as good. But they likely will take longer. It still takes at least a 10 liter one. If he didn’t get a perfect run, we should think 25 to 50 liter fermenter.
If you search on
“50-liter fermenter” anthrax
You get some interesting hits. That is what they had at Dugway in Utah for Project Bacchus.
http://www.pbs.org/wgbh/nova/transcripts/2815bioterror.html
- Mr. Lake is mistaken.
Gerry Andrews, Ivins’ supervisor between 2000 and 2003, is not under a gag order and has addressed all these issues as quoted above and elsewhere.
Two other supervisors Jeffrey Adamovicz and Dr. W. Russell Byrne, also disagree and have spoken publicly on these issues. Jeffrey Adamovicz worked with him for more than 12 years. Dr. A was his boss from 2003 to 2004. He remembers the day the scientists opened that envelope, placed in a double-sealed bag inside a protective hood designed to deal with dangerous pathogens. “The anthrax was floating around inside the bag,” Mr. Adamovicz said. “It was very scary.” He said he turned to Dr. Ivins and said, “That stuff is amazing.” “Yes, it is unbelievable,” he recalled Dr. Ivins replying. “I have never seen anything like that.”
Dr. Ivins, his colleagues said, argued that al Qaeda was responsible. “He was very passionate about this,” former boss Jeffrey Adamovicz said.
None of these men are under a gag order contrary to what Mr. Lake suggests. Mr. Lake apparently has never contacted them to be apprised of the facts. He instead offers his own opinion even though he is not even a microbiologist.
- Old Atlantic Lighthouse said…
- Ed, there was some prior discussion on the number of square feet of plates it would take.
http://anthraxvaccine.blogspot.com/2008/09/comments-by-professor-sergey-popov-on.html
Based on the paper cited above,
http://oai.dtic.mil/oai/oai?&verb=getRecord&metadataPrefix=html&identifier=ADA426293
http://www.dtic.mil/cgi-bin/GetTRDoc?AD=ADA426293&Location=U2&doc=GetTRDoc.pdf
I came to some estimates of square feet of plates from 15 square feet on up. The paper did runs with plates and you can get some numbers there and in the prior posts on this subject.
http://anthraxvaccine.blogspot.com/2008/09/additional-comments-by-dr-popov-on.html
Popov estimates at least 100 plates per letter.
Plates take up too much space and in these volumes are a lot of work. Plus you have to have that many plates. It then takes a long time to work the plates at each stage.
…
There is also the activation time or dormancy time or time to first generation. I have searched on various terms. There are many papers on this, some sponsored by the FBI. The following search gets many interesting hits.
bacillus spores “time to germinate”
One can substitute or try phrases like anthracis, “germination time”, dormancy, endospore dormancy
“spores remained dormant”
and so on. The articles I found didn’t seem to hit spot on the issue of how long until exponential growth starts from the time you take spores and put them in the growth medium.
If we go back to the paper I linked to further above that did the trials of growth times, some runs were duds. It may be that it takes hours or even days to start.
Also as the container is saturated, it should slow. Hypothesis:
dN = k N(Nmax – N) dt
Nmax is the maximum number of spores the growth medium in a flask can support. If he spiked a flask with more starting spores, then he will have a slower growth rate than the average from the paper I cited which started with 1 ml of spore solution per liter of growth medium.
- Old Atlantic Lighthouse said…
- Ed, your point about the 10 percent or so of the first letters being actual spores is important. I was thinking this might be an issue, but didn’t have time to research it or factor it in. I believe Dr. Popov mentioned something in his discussion, but I am not sure if he had a specific number. Can you document the source on the 10 percent figure?
I admire your taking on all comers in a spirited debate. We can learn a great deal from such a discussion and learning science this way is in general valuable.
==Rougher notes which may have repetition of the above
http://www.ncbi.nlm.nih.gov/pubmed/18984151
http://www3.interscience.wiley.com/journal/120832256/abstract?CRETRY=1&SRETRY=0
http://www3.interscience.wiley.com/journal/120832256/abstract
“
Early administration of antibiotics after
exposure is essential – with inhaled anthrax, a delay of just a
few hours is enough to significantly reduce the chances of
survival. Because spores can take a long time to germinate,
antibiotic treatment should continue for 60 days.
Source: JAMA, 1999. 281, 1735-1745.
“
http://www.parliament.uk/post/pn166.pdf
bacillus spores “time to germinate”
http://www.sciencedaily.com/releases/2007/06/070604164928.htm
“
The research is funded by LLNL’s Laboratory Directed Research and Development program, the Defense Advanced Research Project Agency (DARPA) and the Federal Bureau of Investigation.
“
https://www-pls.llnl.gov/?url=about_pls-scientific_staff-malkin_a
https://www-pls.llnl.gov/?url=about_pls-scientific_staff-malkin_a
http://www.pnas.org/content/104/23/9644.full.pdf
“
Spore and Vegetative Cell Preparation: Purification and Germination
Conditions. B. atrophaeus (ATCC 9372) was obtained from the
American Type Culture Collection (ATCC, Manassas, VA).
Spores were produced by using Schaeffer’s sporulation medium
and purified as described (36). Spore germination was
induced by the addition of 100 mM L-alanine, 1.65 mM
L-asparagine, 2.8 mM D-glucose, 2.8 mM D-fructose, 5 mM
potassium chloride, and 25 mM TrisHCl buffer, pH 8.0. The
germination time course was characterized by phase contrast
microscopy before performing AFM imaging. More than 95%
of spores turned phase dark with
“
http://www.pnas.org/content/104/23/9644.full.pdf
page 5
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1877984
http://www.ncbi.nlm.nih.gov/pubmed/17007517
“
The time-to-germination (t(germ)), defined as the time required for the CaDPA band intensity to decrease to the midpoint from its initial value, was found to be stochastic for individual spores with a typical value of approximately 30 min under the experimental conditions.
“
http://www.ncbi.nlm.nih.gov/pubmed/17007517
bacillus “time-to-germination”
http://mic.sgmjournals.org/cgi/content/full/148/7/2089
http://www.ncbi.nlm.nih.gov/pubmed/11964118?dopt=Abstract
bacillus spore “germination time”
activation
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1539616
http://bodyandhealth.canada.com/channel_section_details.asp?text_id=1484&channel_id=1020&relation_id=10875
endospore dormancy
“spores remained dormant”
http://www3.interscience.wiley.com/journal/119420121/abstract
==
Ed, there was some prior discussion on the number of square feet of plates it would take.
http://anthraxvaccine.blogspot.com/2008/09/comments-by-professor-sergey-popov-on.html
Based on the paper cited above,
http://oai.dtic.mil/oai/oai?&verb=getRecord&metadataPrefix=html&identifier=ADA426293
http://www.dtic.mil/cgi-bin/GetTRDoc?AD=ADA426293&Location=U2&doc=GetTRDoc.pdf
==
The actual spores in the first letters may have been mostly grown as opposed to mostly started with. That means the start time was several days earlier. If it was in the middle of a run that takes 7 to 14 days it might have started from before to after 9/11.
==
The FBI in its released warrants or affidavits or its news conferences made it seem like the time he spent in the BSL3 was the conclusive proof he did it. Daschle said so and the FBI briefed him special. The FBI didn’t say it had secret evidence of his guilt and that the time in the BSL3 was not a major part of the evidence.
According to the FBI, Ivins was in the BSL3 2 hours 15 minutes each day Friday, Sat, Sun after 9/11. That is the amount of time he would have had. It may be that he was in and out. Could he go from spores to cells back to spores to envelopes and clean everything and dismantle all makeshift equipment in that time?
If he had 5 liters of solution
on Saturday, how could he reduce it in time to powder spores? (The evidence from above is that there was an attempt to grow a lot. Note there were 5 letters in the first mailing according to current thinking according to the Wiki article.)
1) Build his own lyophilizer and dismantle it when finished.
2) Use a large capacity lyophilizer at the lab. (This seems out.)
3) Spray dryer. Likely not available.
4) Speed vac. Probably too small and might run too long?
Think about boiling a one quart pot of water on the stove until its dry. How long does that take? Note that boiling water with anthrax for over 30 minutes will likely kill it all according to sources that I found. Heating to 40 deg C without a speed vac or hood or special equipment to remove the vapor would likely take too long. There is always the contamination issue of the lab as well to consider.
He would presumably centrifuge the solution first, if he had sufficient centrifuge capacity. He then has to clean the centrifuge. All the equipment he uses has to be cleaned to avoid detection. He has to wear his suit while doing cleaning and every other step. That makes it take longer.
Does it make sense to try to make 5 grams of anthrax starting with spores in 36 hours with a limit of 2 hours 15 minutes per day in the lab?
==
Let’s recap two competing scenarios, among more I know.
In scenario one, a lab with all the equipment to do 5 to 10 gram powder runs is set up somewhere. Workers can work without having to worry about being discovered. They have the top equipment of every type, large floor shaker incubator, big centrifuges, spray dryers and lyophilizers, test equipment, etc.
This lab is given the order to produce 5 to 10 grams. Its sets that up and starts running. But the other comes in to stop and hand over what they have to someone else.
Then in the second batch, they are given the order to prepare more. Possibly right after this aborted attempt. They just go back to work immediately.
Note that the second letters were mailed Oct 19. If you think about a 14 day run in the paper I linked to above and then days to process the powder, it could take 3 weeks going flat out to produce good powder for the next run.
In the Ivins case, Ivins didn’t have the equipment the first time, nor the second. Ivins never had the good setup described above or in the paper I linked to previously. In the Ivins scenario, he doesn’t start again until 2 weeks later and has only 1 week to do it. So again, he has too little time, can only work part time at it, and doesn’t have the equipment.
In comparing likelihoods, each run is less likely for Ivins. So the total likelihood of Ivins is much lower looked at from this point of view.
==
The reason Friday Sep 14 to Sep 16 is so important is that the FBI presented this as the psychological proof Ivins did it. Daschle was briefed by the FBI and he said he believed because of the nights before the mailing.
But now we are told it wasn’t even possible to prepare the anthrax in the nights before the first mailing. This is presented as if the FBI had never said or implied that. Did they go back and tell Daschle this?
The reason Friday to Sunday is so important is the chance not to be seen. If it takes 3 days (72 hours) or more to let an incubator or fermenter run in the lab, then it has to run during a weekday, Friday or Monday and possibly more.
That means others can see it. We are talking about preparing 5 grams or more in a run. That is a bigger run of anthrax than has been done at Ft. Detrick in the recent era. So it would be noticed. It would be a giant run relative to normal.
Moreover, the stats on the paper I showed show that to get 5 grams of powder one should really aim for more like 25 liters to 50 liters because its unlikely that a liter will produce a gram of powder.
The strongest evidence connecting Ivins to the mailings is the lab time. That’s the FBI crown jewel of evidence. But if in fact, he would have to let a large run of 5 liters or more in a flask run (or hundreds of plates) then it is not such great evidence because it would have been noticed.
Note also for 5 grams using Dr. Popov’s 100 plates per letter estimate we get 500 plates. 1 minute per plate comes to 500 minutes, longer than a 2 hour 15 minute period. So he didn’t have time to manipulate 100’s of plates in each step, set up, drying and scraping. Remember, you wanted to have the plates scraped. Well scraping 500 plates takes more than 2 hours 15 minutes.
That the first letters show an attempt to grow shows he needed to. That means the anthrax of the second letters had to be grown. But that means running on week days.
The first letters were unsatisfactory. So new anthrax had to be grown. If it takes 72 hours, then he had to run on a weekday, either with 200 or more plates for 2 grams (if not 400 for 4 grams) or 2 to 20 liters of material in a rotating shaking fermenter. If he used a flask only, his runtime would be longer.
To get 2 grams of powder would likely take 7 days to 14 days not 3. Then it has to be dried.
Air drying will take time. He doesn’t have that on the first weekend. If he air dries he has to stop on Saturday. In which case, how did he expect to get 5 grams to grow if he was going to stop on Saturday to air dry?
Some Ivins emails have been released. Some contain extensive statistical output before 9/11/2001. So he knew how to count. They have tests of statistical significance. So he understood variability. The paper I cited before shows many runs are duds. Ivins knew that.
The FBI used the times in the lab before the mailing as its proof for Daschle. But those times are not adequate to do all the steps. So the FBI case is gone. What is the evidence left?
==
Ed, try commenting on following.
http://www.dtic.mil/cgi-bin/GetTRDoc?AD=ADA426293&Location=U2&doc=GetTRDoc.pdf
Page 12 of pdf methodology. Page 20 table of results.
Note that both plates (NSM) and liquid used an incubator. In each case, 3 days was the best time. But some runs were duds even after 10 to 14 days.
The person thought they could grow at least 5 grams it appears, but stopped with only .5 grams for the first letters, relying on Ed’s linked source and a guess of 5 grams material. But maybe its 10 grams and 1 gram?
It appears someone expected to run 3 to 14 days and harvest a full set. Then they had to stop and give what they had to someone or use it themselves. Then immediately after that, i.e. around Sep 18 they started on the run for the second set.
Did Ivins have a centrifuge to handle 5 liters or more? The paper talks about centrifuging multiple times, each time taking 30 minutes. If Ivins had 2 hours 15 minutes max on each lab day in the first set, he doesn’t have time to centrifuge the 5 liters even once if he was limited to one liter per centrifuge run. Maybe his centrifuge was even smaller than in the paper. Note the paper has a floor size shaker incubator for the liquid runs which weights 400 pounds. The paper had much better equipment than Ivins had. That means Ivins run times would be much longer and his chances to have dud runs or low yield runs much higher.
From the first run having only 10 percent spores according to Ed’s source, we can infer that they needed to grow. That means the second run had to be grown between Sep 18 and Oct 6-9.
If those took 7 to 14 day runs plus centrifuging multiple times, etc. then the work likely started right after Sep 18. So it wasn’t Ivins.
It seems on the first run, someone was in a hurry to dry it. This is more consistent with someone who unexpectedly had to deal with a short time horizon. If Ivins had a small centrifuge for smaller jobs than one liter than he couldn’t have really expected to dry 5 liters or more. So he wouldn’t have tried at all.
Dont’ forget the subtilis that was different from Ivins lab. It is likely a trace of subtilis was present at the start of the run and that this grew along with the anthracis. That argues that a smaller amount of anthracis was used to start and that there had already been several days of growth, i.e. it wasn’t a run from Sep 14 to Sep 15/16 but a run somewhere else started more like Sep 11 or even a few days before.
==
http://www.druckerco.com/drucker/centrifuges/horizontal.asp
Top of line unit appears to be 300ml capacity. The dtic paper did 30 minute centrifuge runs and did multiple ones. So for 5 liters, which is the minimum size run he was doing, it would take 6.5 hours. But he was only in the lab 2 hours 15 minutes each day the first weekend. So there wasn’t enough time to centrifuge even once.
In a BSL3 lab, you don’t want giant centifuges. When you have to clean it, and its when not if, you want as small a one as possible. Also you can’t throw it out instead of cleaning it. Everything has to be cleaned before being thrown out. You want small equipment that is robust and won’t break, not giant centifuges that are going to have repair and break down problems and need a technician to repair who is an expert on that unit but unused to the BSL3.
==
The dog that didn’t bark in the night. Ivins spent only 2 hours 15 minutes each on Fri Sep 14 to Sun Sep 16 by FBI’s own affidavits.
The material that was sent was not sufficiently centrifuged and washed to separate the spores from the rest. This implies the person who sent it had run out of time.
But Ivins had loads of extra time. If Ivins had a centrifuge on those days, then he could have spent extra hours centrifuging and washing. If he didn’t have a centrifuge in his lab, its hard to see how he ever did it.
It appears one of the letters was more thoroughly prepared to cause inhalation anthrax. So if Ivins was doing this and did that one better, why didn’t he do another one better too? He had the time. Lots of it. But he didn’t. This is the dog that didn’t bark in the night.
Ivins didn’t do the Florida run and the New York runs at the same time on this weekend. But that was the FBI’s whole argument. Moreover, Ivins had time that last weekend to do at least one letter better but didn’t. So he wasn’t in there just doing the letters.
==
Lets not forget the subtilis contamination. That indicates a test run was done with the equipment first. That equipment was at least rinsed presumably. Then the subtilis had a growth phase is likely. So there was probably a growth phase of at least a couple days and possibly 7 days.
If we look at the paper I reference before some runs go 14 days. They tested the runs at the end of each day and stopped when they reached a conversion of food to spores.
The people doing the run may have counted on 3 to 7 days and then they got a slow run and they went with what they had. So they started say Sep 11 to mail Sep 17/18 and then got to Sep 16 or so and went with what they had.
One theory (Ross Getman or Kenneth Dillon?) is that there was a transfer in Maine by Atta to someone else that took place on Sep 11. That would then explain their hoping to get a good run before Sep 17/18 but it not working out as shown by many runs in the dtic paper that don’t.
So one scenario is the subtilis prep run was done before Sep 11. That ran in under 7 days. Then that group got the anthracis from Atta who got it from someone who got it from a US site that had it. Then the anthracis run was slow so they went with what they had. Then they started their next run on Sep 18 and got it ready to mail Oct 6-9.
==Comment posted at following thread at Meryl Nass
http://anthraxvaccine.blogspot.com/2009/01/more-pictures-complete-exosporium.html
This is a good dialogue with good contributions by many. Ed Lake is to be commended for his efforts and taking it in stride as have others.
The more steps, time taken, resources used, ideas exploited, money, etc. the more it indicates it was many who did it and not Ivins. Its clear that the silicon fits that.
One test is to look for spores with silicon in the BSL3 or look for silicon on equipment that might have been taken out. If he had leakage into BSL3 in processing then there should have been spores in BSL3 with silicon. Or there should have been silicon left from the processing. This could be under equipment, desks, etc.
Also if he took a lot of spores from the RMR1029 flask to speed up production, used silicon in processing and then put some spores back in RMR1029 to tank it back up, then there would be silicon back in RMR1029. It appears there is none. That means he likely didn’t add back to RMR1029. That means his ability to use a large starting amount from it is less.
This means he needed more growing time to produce his runs. This is very important for both sets, the Sep 17-18 and the Oct 6-9 mailings. The former was on a tight schedule and did have .5 grams of bacillus anthracis spores according to some estimates.
The second run had at least 2 grams of spores. So he had to grow this and he couldn’t speed up the growing by using more from the flask and then returning it. That is very important. Because it means he needed longer run times.
http://oai.dtic.mil/oai/oai?verb=getRecord&metadataPrefix=html&identifier=ADA426293
Carey, Laurie F. ; St. Amant, Diane C. ; Guelta, Mark A.
Is the paper with times for production.
For an external party, not Ivins, putting spores back in their source would not be an issue and we don’t have their flask. For them it would make sense to use silicon to grow. Ivins, however had reasons not to. One is returning to the flask. The other is it would leave a signature of spores in BSL3 that could be detected and finger him. But it appears no such spores have been found in BSL3. It seems almost impossible that Ivins worked quickly to do these large volume runs without leaving spores in the lab containing the same silicon. Thus this is another proof that Ivins didn’t do it.
The subtilis contaminant is another proof he didn’t start with more from the RMR1029 and then put some back. So we have to count on longer runs from smaller numbers of starting spores if Ivins did it. Together with the dtic paper linked to above, that implies Ivins didn’t have the time is the reasonable interpretation of the evidence.
Also there should be some subtilis spores in the BSL3 if he prepared it there. Some of those should have silicon too it would seem. Those also were not found and that’s more evidence it didn’t happen there.
==
Given the large production volumes and the short time and ad-hoc equipment that was likely not up to the volumes done so requiring mutiple use, there should have been spores of anthracis and subtilis spiked with silicon all over BSL3. None have been found. That is pretty strong evidence the processing didn’t happen there. Not with make shift methods of drying, processing, etc. Not with feet of plates all over the place if he used plates. Not with drying that didn’t have an adequate hood. Etc. Etc.
==
Also the silicon was in both runs. So it should have been
in the BSL3. Since the silicon was in both runs, his ability
to start with a larger amount to speed up a run wasn’t possible
with either run. So that method is subject to a sum inequality,
i.e. the amount taken out from his flask in sum was limited
by whatever bound we have on the decrease, at most the flask
size minus the amount in it being one estimate. (There is
still remove and add water to consider?)
His better method was to start with more from the RMR1029 flask not
use silicon and then put back some into his flask. Both times
that was his better method. After the first failure, he should
have tried that the second time.
For an outside scenario, they could start the second run
on Sep 18 and get it done in time and do a precision job,
taking as long as needed for each step to get it right this
time. The date Oct 6-9 when mailed was when it was ready.
Ivins could never feel such freedom.
The fact of a second run
with everything done right by itself is unlikely for Ivins
since he would never have the feeling that he could afford
to do each step of the procedure until that step was done
right and then pick up the next step.
In fact, each step would
have to end or be interrupted for the need of concealment.
His lab times are insufficient for each step to be done
perfectly. He couldn’t produce a perfect run with a start stop
limitation over the last week, Friday to Friday.
==
If the spores he cleaned up in his work area outside BSL3 had silicon in them, shouldn’t there be some traces still?
