Oliver Smithies:[00:00:29] This is book six, Roman six. Covers the period June 1955 through January 1956. Still at Connaught Medical Research Laboratory. It’s a straightforward beginning, Thursday June 16th, a family study checking on their assignment to the F1, F2 and M designations of their haptoglobin types. Page five and six, a summary [00:01:00] of several families using the notation of one for what was originally M, 2A for what was originally F1 and 2B for what was originally F2 that are here. [name redacted] family, [name redacted] family, [name redacted], [name redacted], [name redacted], [name redacted], [name redacted] and [name redacted] families. More families recorded on pages nine and 10 with a comment that later appears in the summary. The frequencies of the normal adults fully checked unrelated by birth or marriage [00:02:00] are the males were six, nine and four in the three types, type one, type 2A, type 2B, and the females were type one, 2A and 2B in the ratio two to 12 to eight. Total was 41 and the P value was P as the frequency of one of the genes — no, I don’t know what — hold it. This was assuming that they were homozygous for type one and homozygous for the elderly type two, that P and Q would therefore be .45 and .55. The frequencies were [00:03:00] six, nine, four, two, twelve, eight. No, that’s wrong.
— with a note that the frequencies of the normal adults fully checked, unrelated by birth or marriage were males six, nine, four and one, 2A and 2B. Females two, twelve and eight in the same category. A total of 41 and a P value, I think it’s probably related to assuming they have the same relative frequencies of the .195 for one, .512 for the other and .293 for the third. The P and Q are the frequencies of two possible genotypes. Page nine also has frequencies [00:01:00] of the marriages expected from these gene frequencies. Pages 11 and 12 are continuing with family studies of the [name redacted] family. They were bled on June 21st, but evidently some problem here, maybe because it says on page 11 that all were misread as F1 on the hemoglobin check but were found to be two F2 on checking against the [name redacted] family. So I must watch to cross check with standard serums very carefully. In fact, must run at least one member of the family against standards.[00:02:00] Page 15 and 16 shows some of these checks with the [name redacted] family and all checks are OK and up-to-date. Continuing June 27th Monday, nothing particularly notable there. They had Japanese families noted as mother and father, M and F1. [name redacted] family June 28th, page 22, [name redacted] family checks, page 26 on July sixth. With a comment here that [00:03:00] Dr. Walker, that’s Norma Ford Walker, reports mother corn on blood, difficult on blood route, reasons to be discussed. Friday, July eighth accrued rapidly antiserum to molycono serum to be made for a rough test on a [00:04:00] large rabbit. Was given 1.5 to two mils of serum intravenously. It’s a very high dose, but a single shot is to be tried and bleed in 11 days. And with a comment that a sequel bled on Thursday, July 21st, 13 days oodan test set up Friday 22nd of the results. Monday 25th a very heavy precipitate so that I got an antibody against molycono serum. A second sample from molycono was tested on July 12th and with an adopted child. [00:05:00] And here I tested the gel slice with a histochemical test, Seligman et al’s test for cholinesterase, so this is the first record of my doing any experiments related to enzymatic activity and this [00:06:00] later become a very important use for the starch gels to do enzymograms, as one might call them, for looking at where various enzymes were in the sample. Starch was good for this because it was very biologically compatible and didn’t have any of the free radicals that are present in the later gel material, polyacrylamide, and so was very amenable to testing enzymatic activity. And this was a diazo blue test and six bromocarbo naphthal nepsoxycholin. So it was a test for cholinesterase. The activity is in the beta fraction is the [00:07:00] comment. That’s on page 30. On page 29 there’s a comment that this is not so, see the references. They claim that their red blood cell cholinesterase does not attack carboxy — let me try that again. The gels were sliced and tested with this dye for cholinesterase and I found the activity in the beta fraction. And I thought at the time that it was red cell cholinesterase, but on the facing page I say this is not so as they claim that the RBC cholinesterase does not attack carbonapsoxy choline, so it’s not [00:08:00] from the red cell. So I became interested in this topic and so prepare tomorrow hemoglobin three OS serum plus if possible sample from Otto and read up specific inhibitors and test for plus or minus fresh hemoglobin. Use the best technique, starch technique, for full scale repeat on one of each of my stored samples of different types. [00:09:00] And on Wednesday, July 13th, page 32, I got the bloods from infants to be used partially for cholinesterase tomorrow. Nice comment here on page 32 and 31. The serum prepared Thursday very carefully without preliminary centrifugation and the serum was limpid and absolutely free from yellow color. Odd comment. The blood from infants. They were infants, four weeks, six weeks and two and a half months, so free from yellow [00:10:00] means presumably no bilirubin or whatever. Thursday, July 14th, using the bottom half of the gels for cholinesterase and the top half for ordinary staining. And then on page 36 and 35 I have an insert showing the location of the beta naphthal liberation as I call it, the cholinesterase activity and the staining pattern, to show where the main staining is. I haven’t yet understood [00:11:00] what this is going to lead to, but it’s going to lead eventually to understanding that the gel separates proteins by size as well as by charge so that bands occur in different places when looked at in free electrophoresis, for example, on filter paper versus an electrophoresis in starch. But there’s a very nice diagram of where beta naphthal color came, which is cholinesterase, versus the protein staining.
This is a test with beta galactosidase on page 36 and it says [00:12:00] all are negative after 60 minutes. Beta galactosidase all negative. I don’t quite understand why I’m — this is a test for the naphthal liberation, not for beta galactosidase. The beta galactosidase all are negative is an addition in red crayon, so it’s a comment afterwards. The gel results are of naphthoxy naphthal. [00:13:00] Beta naphthal liberation from naphthoxy choline versus beta galactosidase is a completely different enter. For Friday, June 15th, I repeated the samples, nothing remarkable. Here is enzyme localization. Beginning to understand that it’s necessary to find the localization on filter paper as representing free electrophoresis versus the location of the bands in starch gel. So [00:14:00] page 40 is the beginning of this type of work, particularly encouraged by David Poulik who was interested in this sort of problem. So two filter paper samples dried at room temperature were stained with bromophenol blue and tested for cholinesterase. And there is a little diagram saying that the enzyme is equivalent to the back of alpha one even when counterstained, not beta because I expected it to be a beta, not alpha. So here is albumin alpha one, alpha two, beta and gamma and on the [00:15:00] filter paper and the enzyme appears to be the back of alpha one even though my test thought it was beta because of the uncertain notation of my bands versus the bands that are seen in free electrophoresis. It took a while to work this out and it’s actually documented rather well in another article that might be looked at in conjunction with these books that I published on the history of gel electrophoresis. I can supply the reference later.
Let’s just pick it up where it was. Do you get the —
I see it. It’s ready.
Is it ready? Can you go back a little bit and let me just see it? [00:16:00]
If we go back —
An account of that early history of gel electrophoresis can be found in a chapter that I wrote for a journal called Methods in Molecular Biology published by Clifton, New Jersey in volume 869, pages one through 21, and it’s titled How It All Began, A Personal History of Gel Electrophoresis. So it won’t make sense as we go through the pages of the notebook as well as it will if a person wants to get that history by going back to that more rational account. So here we go anyway. This was [00:17:00] the beginning of looking at enzymograms with Dave Poulik initially and I ran a serum sample on the infant 832 on page 41 and 42 of this book and determined the position of the sample using the substrate. Cholinesterase substrate is all I record it as and I could clearly see that the [00:18:00] enzyme stained between alpha two and betaglobin using the notation of albumin alpha one, alpha two, beta and gamma globulin, which is what we used at that time. So here was the sample. Very, very clear examples. And then I have on page 43 and 44 some of Dave Poulik’s serum testing the same — using the same method. With a comment that [00:19:00] it’s clear that the enzyme can migrate close to beta or close to alpha two or intermediate according to the conditions. So there were slightly different conditions where the enzyme band was slightly variable according to the particular electrophoresis that was done. But it was somewhere between alpha two and betaglobin.
More tests on July the 18th with sample 830 again. That’s the infant sample, I believe, an infant sample, but this was 830. [00:20:00] There were three infants, 830, 831 and 832 and this was set up 830 again and all three infants. And rather a nice diagram on page 45 showing where the esterase bands were relative to the bromophenol blue stained gel. There was an absence of substrate. There was no band there. And the conclusion is that a major enzyme is ahead of beta slower than alpha approximately equivalent to the filter paper method because [00:21:00] I’m still thinking in terms of mistaken identification of the various bands. Now Tuesday, July 19th I began to take samples directly from the filter paper and put them into a starch gel alongside normal serum in order to compare directly where the bands are. Now the protein level, not enzymatic particularly. So Tuesday, July 19th is a beginning of that sort of experiment where I took Dave Poulik’s serum [00:22:00] that was run on the filter paper and then three samples were cut out which are labeled alpha one, alpha two and beta globulin according to the notation used for serum. And then they were put into the gel alongside the whole serum. The results are very poor, it says, on serum. Bromophenol blue wrecks things. However, the ones with beta filter paper give clear bands lining up with alpha two. So what was beta on the filter paper is lining up with what I thought was alpha two, very much underlined. So that the two things don’t coincide. The beta lines up with [00:23:00] alpha two. That’s what I called alpha two in the starch gel is a beta in filter paper. Beginning to see the conflict of the two methods. That’s not to say conflict, but the difference of the two methods.
July 20th, Wednesday, page 52, 51, similar experiments cutting out alpha two, beta and gamma from the filter paper and albumin and alpha two samples or albumin one of alpha one from filter paper and running them alongside serum samples in the gel. [00:24:00] Still getting still more complicated by page 53, 54 when now I’ve taken six samples across the region of alpha two and beta in the filter paper. I’m looking at filter paper samples covering the range of alpha two to alpha one, six samples put in alongside — into a starch gel alongside serum samples and put it as fully checked with a comment that post alpha two extends over all the filter paper, alpha two and back to beta, etc., etc. So that some things fit and some things didn’t fit. Beta [00:25:00] filter paper is equivalent to alpha two in starch. So conflicts, they’re not the same.
Friday the 22nd, page 55 and 56, again in this case 10 samples along the filter paper, electrophoresis, compared quite laboriously with a gel for each sample. So that’s basically 10 electrophoresis samples to do this analysis and 10 bits of gel. It looks like the hemoglobin band of protein are alpha two and beta are equivalent to the old alpha two, etc., etc. Confusing, confusing. Even more extreme by now, July 26th, page 60 [00:26:00] and page 59, here there are 20 samples cut from the filter paper going all the way from gamma up to just short of albumin with a gel for each one. Laborious, laborious. Comments on page 62 that fully checked the paper, band, etc., etc., and a long list of what was being found. More samples of this sort, 20 samples on page 63 cut from the filter paper and compared in 17 different gels. Very laborious. Trying to make sense [00:27:00] of it. For example, alpha one very clear first expose of it. Maybe this one is a better example. Number 15, sample 15, was beta on the filter paper, very clearly beta on the filter paper, but it’s very clearly alpha two in the gel.
Continuing this laborious work, one through 20 samples cut from a single filter paper strip on page 65, 66, interpretations of it. And trying a little bit more to understand alpha one on page 67 [00:28:00] and 68. With a little relief on page 69 and 70 where I’m doing a simple sampling from a family in gel compared and confirmed on page 72. Beginning to think of the frequencies of mating, of whether they occur approximately as expected, seen on page 74. More family page 76. Now going to enzymes again, beta naphthal acetate August 30th and 31st, page 78. Getting a very clear enzyme band. And [00:29:00] showing that the beta naphthal acetate is very much like cholinesterase. Family study conjoined male twins, difficult family. Starch test September 13th, 20%, 70%, 25%. Back to filter papers again, page 83. A person with nephrosis. I’m sure this was in conjunction with Dave Poulik. Mrs. [name redacted], [00:30:00] high alpha two. She had kidney problems and so there’s a very, very heavily staining filter paper band which is alpha two and I cut it into six samples and ran them against a starch. I don’t see the results of it here, but no doubt it’ll turn up. Oddly enough, I don’t see the conclusion of this experiment.[00:31:00] Wednesday, September 14th just shows what was set up, but it doesn’t describe the results clearly. But she was with hemoglobin I had the comment unrecognizable by hemoglobin. So she doesn’t have a clear hemoglobin binding band and a very questionable 2B by stain. It’s what I would now be fairly sure is that her alpha two is what in the starch gel I call slow alpha two. It’s a macroglobulin that stays in plasma when an individual [00:32:00] has nephrosis. Mrs. [name redacted]. And my study with [name redacted]’s family, history of two infants, both badly deformed and an attempt to see whether there was some association with RH and Norma Ford Walker is looking at the female that’s rH negative, a possible history of malformation in the family. Both parents and late children. [00:33:00] Not very clear.
Looking at Mrs. Skono’s rabbit serum on Friday, September 16th, and four guinea pigs were also injected with molyconos serum again, also with my serum and Dr. Scott’s DAS serum and Dave Poulik’s DP serum, so we had four guinea pigs with each getting different samples of serum. Some problems again looking at samples of [00:34:00] families that have troubles. Rabbit given serum. Looked at a carcinoma sample, a lung carcinoma which had been taken by Dave Poulik and should have increased alpha two is the comment. More family studies. October 28th, page 98, samples from 13 Eskimo [00:35:00] people dispatched 26th of September, 1955 and I’m looking at them October 28th. These samples stood in the cold for over a month and they’re very badly hemolyzed, for the most part de-bred, and ran them. But I could type them all for the haptoglobin types, one, 2A, 2B.[00:36:00] More families. November first, more families. Beginning to look at some antibody results with rabbit antiserum on page 101, rabbit antiserum and type one, type 2A, various samples. Wednesday, November second, page 104, I took samples from a McLean serum which is high in alpha [00:37:00] one. I cut samples and ran the usual comparative with starch, with a whole serum in starch gel. And on the opposite page, page 103, I began to do some auto centrifuge work with totto serum. Looked to be not very good resolution in the auto centrifuge from the pattern that I sketched, but there was some attempt to measure the [00:38:01] sedimentation of albumin and globulin, etc., in a rather complicated graph that’s inserted as a loose into the gel — into the book at this point. It would not be in the right place. I’m going to check it. Wednesday, November second, page 104 has a straightforward experiment comparing samples cut from McLean serum covering a little bit behind albumin back to alpha two. Six samples were taken from the filter paper and compared with the whole [00:39:00] serum in the usual way in the gel.
Also on the opposite page is an attempt to look at some ultra centrifuge results where the ultra centrifuge with totto serum showed the usual albumin alpha and gamma globulin, but very fast sedimenting bands that I don’t know why I was interested in at that time, but there they are. November third, pages 105 and 106, some rather ineffectual antibody experiments looked at. [00:40:00] I was obviously following up on the type of things I’d done when I was in Wisconsin with antibodies and rather complex and fairly carefully done, but I don’t remember any good conclusions from these experiments. I was trying to do some [00:41:00] sort of combination of immunoelectrophoresis and using the gel with an antibody and antiserum on top of it, a sort of like people did with photo paper. So I would have starch gel on the bottom and arga and antibody on top with the antiserum and looked to see what bands I got. Page 110 I say that very poor electrophoresis, so do not expect good result although very difficult to set up well. A comment that when I took it off on Thursday, three days later, I dismantled the electrophoresis, I dismantled the arga test and found precipitation bands that had drawn in relation to the gels, about five [00:42:00] bands in the combined starch gel electrophoresis and the immunology. So there’s something I thought was a fast alpha two and beta and a question mark on something gamma, four different precipitation bands. One very much corresponding to what I was calling beta which ended up to be transferring.
What I’ve said, no certain precipitation could be seen with Dave Poulik’s serum, but the zones for OS were very difficult to see and were too flat, much too [00:43:00] much antiserum, so not a very good result. Pages 112 and 111, ring tests against type one, type 2A and type 2B to see if I could see different results. Many, many bands, but not really interpretable. For example, on page 114 I commented it looks as if all the components in type two serum are precipitated by the antibody against one and one antiserum, possibly one extra one, in two that is absent in one. So I’m [00:44:00] trying to look for differences, but it doesn’t look as if I’m being able to find them. Wednesday, November 9th, there’s a male policeman with a cholinesterase abnormality. I suspect that this was got from Verner Kalo who was a colleague in Toronto and then looking at — no, I’m sorry, that’s a different thing. [name redacted] here is a male policeman with a cholinesterase abnormality, presumably from Dave Poulik, to see if I could see anything different. [00:45:00]
And the enzyme locations are sketched on page 118. Main enzymatic activity is the maximum activity is shown as a single band. November the 10th, Thursday, going back to try to get an immunoelectrophoresis working with the starch, with the three types of [00:46:00] serum, one, 2A and 2B and with an antibody. And the following page is very complicated results of the antibody against these things with a comment, it’s a very complex immunoelectrophoresis with a comment that probably the antigenicity of slow alpha two beta, fast alpha two, clearly defined in general. I don’t quite understand it. But here there’s a diagram from [00:47:00] type one, type 2A and type 2B and they really at this point certainly look difficult to distinguish.
A family again, page 124, 123. [00:48:00] I tried some bovine serum on November 15th and also a sample from an individual with multiple myeloma, but 11 bovine serum were run. And I could see differences between the samples. [00:49:00] There’s a sketch there which I later found were differences in transferring, but at this point I obviously don’t know that. I just have a sketch of what the results were showing. There were quite a lot of band differences in the region that I called beta globulin. Beginning of understanding of the casual transference. I think these samples probably came from the person with whom I later collaborated in doing bovine transference, but I don’t have any note here that that’s the case. But I found four different types, one [00:50:00] not the same as samples two and three and not the same as samples four and five and six and seven. So I had four different patterns in these 11 serum. Looking at the person with the cholinesterase abnormality, November 17th, page 130, sketches of the enzyme results of cholinesterase with a new sample.[00:51:00] Good sketches on page 129 of the cholinesterase tests of the enzyme bands versus the protein bands and some comments on immunoelectrophoresis again with a comment that there were no successful results on this. Page 132 and 131, there are some things that are scribbled out, but between those pages is what I think is probably [00:52:00] the results of the cattle serum. It’s showing bands that I’ve labeled as one, two, three, four, five, six bands appearing in what later, as I said, are going to turn out to be transference. They’re very clear cattle sample. I’m expecting, as I said, four groups at the moment. More bovine sample. Guinea pigs being injected with serum, November 22nd, still haven’t [00:53:00] given that up. And here I’ve got on page 138 and 137 the summary of the bovine results. I have all together what I call four groups, various animals in the four groups.
A straightforward family study with a family with a history of abnormal children and miscarriages, trying to find out things that are going wrong. [00:54:00] Trying starch tests on November 29th, Tuesday, corn starch gelatinized at nearly 100 degrees in autoclave, etc. Thursday, November first, working with defatted and dried corn starch. [00:55:00] Trying to separate their different components, different factions, from the starch. Not very practical. A simple family study, December second. Three more families, December fifth, third and fifth, [name redacted] family, [name redacted] family and the [name redacted] family. They had triplets. [00:56:00] These tests went on for several days. I’m still checking the samples on Thursday, December eight, page 158. Then 159 and 160, a summary of total of 19 samples. Father, mother, etc., etc., and which were 2A and which were 2B and which were 2A again. All put onto one gel that was [00:57:00] run, 12 centimeter gel sliced horizontally to compare all of these samples.
December 14th, page 164 and preceding page 163, shows again how laborious these tests were to try to find the equivalents between starch gel, electrophoresis and filter paper. [00:58:00] Here is a rather beautiful sample of filter paper electrophoresis from which 20 fractions, 20 strips, were cut covering albumin all the way up to gamma and then run in a total of 20 different gel patterns alongside serum with a comment that you must read the top and bottom of all slices. Evidently not happy with the result with detailed comments of what was found. We’re getting close to the solution to this problem, but whether we reach it in this book I’m not sure. [00:59:00] Another example on page 168 and 167, just in this case trying to look for the pre-albumins. I’ve got four samples, one, two, three, four, which just look ahead of albumin down to albumin and then five, six, seven, eight which are covering alpha one and sample nine is alpha two and sample 10 is beta on the filter paper notation and then run very clearly on gel of which there were things that were significant as far as the future interpretation are concerned. Samples one, two, three, and four. [01:00:00] Four shows a faint albumin. I can’t really see what’s happening in front of the albumin in these gels and it’s bothering me. I want to find out what’s happening in front of albumin and I can’t really see it. Pre-albumin one, two, three, four, nothing notable.
And then a little interlude on Saturday the 17th, just a family study, and [01:01:00] a filter paper electrophoresis sample. But not any clear notation of what’s happened to that filter paper work. And then on page 171 and 172 I find the way to solve all of these problems of locating the — of comparing filter paper with starch gel and it’s so simple that it’s remarkable. Instead of cutting out the samples from the filter paper, one, two, three, four, five, six, I’m up to 20 going transversally across the filter paper. All I had to do was think in a different frame and cut a strip going all [01:02:00] the way down the filter paper and put it into starch gel and run it at right angles. In other words, a two dimensional electrophoresis. And this result was first got by the two page — I’ll try that again. Pages 171 and 172 in this book are really very important because they show a simple change that makes an enormous difference in being able to understand the differences between filter paper electrophoresis and the starch gel electrophoresis. And it was in trying to [01:03:00] understand the pre-albumin that I’d looked at on pages 168 and 167 where I had four samples in the region of the filter paper ahead of albumin and hadn’t been able to find any proteins there. So on page 171 I realized that I could cut that sample in the opposite direction. So I’m taking a sample from the filter paper that covers the whole of the region from pre-albumin back to albumin and putting that into the gel and running it at right angles. And then by running it at right angle I can see where things come from.
So it was [01:04:00] the invention of two dimensional electrophoresis. So the filter paper electrophoresis has — the one that I have in the example is the whole of the filter paper sample is cut into a single strip covering albumin, alpha one, alpha two, beta and gamma globulin and put into starch gel electrophoresis at the origin and then run at right angles. So it’s the invention of two dimensional starch gel electrophoresis. And it was such a moment of insight that I remember exactly where I was standing. I was next to a centrifuge, which was a critical part of the apparatus that I had, and I suddenly thought if I turn it at right angles I can see [01:05:00] what’s happening. And on page 171 I point out some difficulties. A difficulty in cutting the filter paper, the wet filter paper electrophoresis which I later solved by just making the electrophoresis on strips of filter paper instead of cutting the wet filter paper. And that was one problem, cutting wet filter paper, and the other was that the slit closes up and it’s difficult to get the paper down into the slot, but we use a holder to keep the slit open is my comment. And results quite satisfactory. The beginning of two dimensional electrophoresis.
Page 174 repeats [01:06:00] the pre-albumin experiment so that the filter paper is cut longitudinally including albumin, pre-albumin area and set up at right angles in the gel and also another sample using the whole of the filter paper. So two gels are on pre-albumin and the whole serum with a comment at the end of it that the albumin results are the same as yesterday, serum is not much improved, more protein than previously, but slow alpha two is very streaky. I decided to test a complete range of filter papers to find out if it’s possible to get a filter paper which won’t absorb the protein because I’m noting that this experiment [01:07:00] was done with SS598 paper. That’s Schleicher and Schuell 598 paper, so it’s clear I need different paper. And then so on page 176 I tested several different papers, Whatman 3mm, Whatman three, SS Schleicher Schuell 589, Munktell, Schleicher Schuell 598 and SS589YD which proved to be SS598YD. Transposition of the two digits. Many of them were tried and Whatman number one, a very thick unknown one, Healy, etc. And so the comment is that on page 175 except for the thick unknown and the Schleicher Schuell [01:08:00] 2043B, which are poor, there is little to choose between the papers except for the thick one and one of the Schleicher Schuell ones. SS589 appears slightly better than the others. None show nearly as bad slow alpha two in the two dimensional experiments, so the slow alpha two effect of streaking must be due to the filter paper electrophoresis, possibly heating and denaturation, etc.
Some antiserum tests still going on December 28th and December 29th. Not worth much comment. [01:09:00] Infractionation of the starch, trying a butyl alcohol recrystallized cycle hedging precipitate, etc., etc. It didn’t work out, I don’t think. More important is the two dimensional electrophoresis. I’m back to it on December 29th, page 184, to determine whether the poor slow alpha twos on two dimensions are due to heated denaturation around Dr. Scott’s DAS serum in cold and at room temperature, on Schleicher and Schuell 598 and Whatman number one paper, etc. And the filter paper results are commented on page 183, are that [01:10:00] SS589YD and Whatman number one and XM are no appreciable differences. On the opposite side I say on page 183 filter paper electrophoresis on Schleicher Schuell 589YD are very promising, probably the best of the lot. Certainly better than SS598 and the wet strength is very good and easy to handle. So since the results on two dimension are also good on this paper, try a full experiment on it.[01:11:00] So January 22nd, Monday, an experiment was set up with SS589YD paper run for 13 hours electrophoresis at 180 volt, so many volts per centimeter and then two dimensional gels. I had a comment on this, two gels were run, single paper cut wet, reasonable, but the paper electrophoresis was very poor. And double paper cut, still cutting the paper and difficult to handle. Trying again on page 190. I haven’t yet realized that all I need to do is to use [01:12:00] strips of paper and not try to cut the wet filter paper. Page 189, paper cut wet, cut longitudinally, then transversally to overlap and still have only 7.5 centimeters on each, etc., etc. Not realizing yet how to make it simple. I had a good result on January fourth. Two dimensional. Not much – the filter paper electrophoresis is still not very good. Trying some more of two dimensional experiments. Still cutting wet filter paper. [01:13:00] Saturday, January seventh. Beginning to think about using a different filter paper. Continuing tests of this sort.
And then suddenly on page 204, January ninth, I realized the solution. Use filter paper strips. Don’t cut the filter paper. So Monday, January ninth is a good day in which I say on the opposite page, 203, therefore in future abandon all but XXM filter paper and get conditions right [01:14:00] for that filter paper and using strips of filter paper now, not cutting wet filter paper. Several narrow strips of XX over two papers just to see if a separation is possible this way and so avoid cutting and make the amount of serum more certain. Let me stress that again. Monday, January ninth, page 204, I realized that the best thing to do was use strips of filter paper instead of cutting wet filter paper. Use narrow strips and edge effects then were disappeared and the resolution was good and I could use a filter paper XXM which was sturdy and easy to handle. And so here on page [01:15:00] 206, January 10th, I have an example of some of the strips that were used and stained. And so on XXM paper on page 208, 207, looking at the patterns.
Saturday, January 14th, two more families. Juvenile diabetic family. Bovine serum again on January 17th. I’ve already begun to know how to type them and I typed these samples as [01:16:00] what I call type three on page 217 and type four from previously worked out types. Bovine serum again, January 18th. Began to photograph samples because on page 219 in pen is added, photographed contrast process pen with a red filter and beam illumination, very good results, better than naked eye. First really systematic attempts to begin to think about taking photographs more. Bovine serum, a few samples that had not been correctly [01:17:00] or typed well enough and being compared. One, two and four are the groups. With a possible 2A and 2B. Human cross checks on page 222 and we are going to end this book shortly. Various enzymes being thought of, pancreatic lipase, lipase in esterase, references to these esterases, phosphases, naphthal dicarbon, dicaprylate, naphthal carbonate, naphthal lowerade and – references to these various methods.
And on the last two pages of this book [01:18:00] are bovine serum samples, one, two, three, and four, put into their groups one, two and three, four. New bovine serum the size of the previous lots. With a comment that hemoglobin does not look to form precise bands, rather a general area, but I’m still having no difficulty typing the cattle serum. End of that book.