o
The

Makers of the Revolution in Biology

by HORACE FREELAND JUDSON

Simon and Schuster o New York


30 o The Eighth Day of Creation
four carbon atoms, shares two adjoining carbon corners with the pentagon,
which has another pair of nitrogen atoms in it. Guanine and adenine differ
only in small side groups attached to other corners of the hexagon. These
two bases are called purines because of their chemical relationship to uric
acid and so to urea-and biochemistry is conventionally said to have begun
with the synthesis of urea, by Friedrich Wijhler, in 1828. The three other
bases, thymine, cystosine, and uracil, 1893, 1894, and 1900, are called
pyrimidines, a longer name of unilluminating origin but a smaller, simpler
structure: a single ring, just the same hexagon of two nitrogen and four
carbon atoms-with side groups, again, that make the differences.
By the 192Os, it was realized that there are two kinds of nucleic acid. In
one, now called ribonucleic acid, RNA, the bases are adenine and guanine,
cytosine and uracil. In the other, the ribose sugar lacks a fringe oxygen
atom-hence deoxyribose nucleic acid-and a pyrimidine has been
switched, uracil replaced by thymine. Uracil had first been found in yeast,
and was known in a species of wheat. Thymine had been discovered in calf
thymus gland, whence its name, and was known in every animal cell where
it had been looked for. Uracil and thymine are very similar. For a while it
was thought, then, that ribonucleic acid, bases G A C U, was for plants and
deoxyribonucleic acid, G A C T, was animal. This idea collapsed in the
early thirties under accumulating evidence that both RNA and DNA are
universal. By then, too, it was known that the chromosomes are in large
part DNA. Nonetheless, DNA was thought to be built up in the simplest
way imaginable, with the nucleotides following one another in fixed order
in repeated sets of four. This exceedingly elementary picture was called the
tetranucleotide hypothesis. It was propounded by Phoebus Aaron Levene,
at the Rockefeller Institute for Medical Research, an organic chemist of
highest reputation. Accurate measurement of the proportions of the bases
in samples of DNA was impossible with the chemical techniques available.
And so the belief was held with dogmatic tenacity that the DNA could only
be some sort of structural stiffening, the laundry cardboard in the shirt, the
wooden stretcher behind the Rembrandt, since the genetic material would
have to be protein.
Rigorous proof that the gene is DNA and not protein appeared in 1944,
when Oswald Avery and fellow workers at the Rockefeller Institute in New
York published a paper in The Journal of Experimental Medicine about
inheritable transformations that occur in a strain of pneumonia bacteria
when they are mixed with DNA extracted from a different strain. Avery's
paper is today universally, cited as fundamental, always with the reserva-
tion that the proof took years to be credited. In February 1944, when the
paper appeared, Crick was working for the British Admiralty as a physicist,
designing naval mines, and Watson was a precocious college boy in Chi-
cago, consumed by ornithology the way another might have been absorbed
in railway timetables. When they met, seven years later, both knew of
Avery's work, though then and for several years more it was still generally
believed and widely asserted that genes are protein.
On the point of picking up Avery's paper, I realized that for the moment


                             12  "He was a very remarkable fellow." o 31

I was surfeited with abstractions about DNA. I wanted to know what the
stuff looks like, how it is prepared, something of how one tells that it is not
protein. In the entrance stairwell of Perutz's laboratory building, before a
floor-to-ceiling window, I had stopped to look at a molecular model, eight
feet tall, of the double helix of DNA, and stopped again, on the landing
above, at a model, nearly as large, that claimed it was of the alpha helix, a
structure in proteins discovered by Linus Pauling. Surely, I thought, the
two models could be told apart if one knew what to look for. The stretch of
protein wound up and around, linked back and forth to itself by banisters
that were jarringly off the perpendicular, bobbing and weaving upwards in
a boxer's shuffling syncopated rhythm, while the DNA, right enough, was
double, the strands separated by alternating narrow and wide grooves con-
nected not by balusters but by the horizontal bases, planes laid flat to form
a slowly revolving series. One could say the DNA had a calmer, cooler
architecture. But really it was preposterous, despite the monumental use
to which these models had been put, to think of them as aesthetic objects.
The overwhelming impression that each gave was much the same: a vi-
sually confusing lacework of thin rods intersecting at knobs colored var-
iously blue, red, black, white. This DNA was still an abstraction. Feeling
exceedingly elementary, I asked Sidney Altman, a friend and molecular
biologist now at Yale, to show me what DNA really is.
The afternoon I arrived at Altman's laboratory, he produced a large bottle
of a gray liquid that looked like dishwater. "We start with this. Bacteria in
a culture medium-just water, some salts and a carbon source, and some
amino acids they need. They've been growing since I started them last
night. That's why it's so cloudy: you're seeing a mist of bacteria-not the
bacteria themselves but the turbidity they cause by scattering the light.
You realize, the chief thing you're going to learn is how easy things are
once you know how to do them. Now we centrifuge them down, to concen-
trate the bacteria." He poured the gray liquid into four stubby plastic test
tubes,. which we took over to a large box, like a stainless-steel washing
machine, with a top hatch that lifted heavily. Wjthin was a pear-shaped
spindle with slanting holes like spokes. Altman put the tubes into the holes
and set the machine's speed control at 8,000 rpm. "Come back in half an
hour."
When Altman took out the tubes, the liquid had turned water-clear, while
at the bottom of each was a small heap, in color the pale yellow-gray of an
old nylon shirt. "This is ersatz science," he said. "We're not doing this for
any real experimental purpose. Takes the edge off one's precision. The cells
are all in those pellets. We can throw out the supernatant." He poured away
most of the liquid. Then at his bench, he scraped up the pellets of bacteria
with the end of a glass rod and transferred them all to one tube with a little
liquid. "They're back in suspension but concentrated a hundredfold." In-
deed, the liquid, clear a moment earlier, looked filthy. He turned to an
appliance the`size and shape of a melon, with a large rubber navel on top
that began to shake with silent laughter as he pressed the end of the tube
to it. "Breaks up the clumps." In a burlesque of the classic gesture of the


                           P
32 o The Eighth Day of Creation
movie scientist, Altman held up the tube to the light as he added a liquid
from another bottle. "This is EDTA-ethylenediaminetetra-acetate-which
takes up the magnesium from the bacterial cell walls. Bacteria have tough
cell walls, but this stuff makes them very weak." He searched a shelf,
found a plastic jug labelled "10% SDS," added some of that. "Sodium do-
decyl sulfate- all it is is a detergent; you could wash dishes with it. We put
it in to solubilize the cell walls. The idea is to break the cell walls to get the
DNA out." The technical term for rupturing cell walls is "Iysis." To lyse
cells, biologists use strong chemicals, or even grind the cells in a mortar
with sand; animals get the same results with the subtler means of an en-
zyme, called lysozyme, present in such body fluids as tears, saliva, and
intestinal mucus, which has the protective effect of breaking open bacteria
that attempt to invade. Altman put a black rubber stopper into the tube.
There was about a quarter cupful of liquid in it. "As the cells lyse, the
bacteria will vanish and the mixture will begin to clear. At the same time it
will get very viscous, because the DNA in each cell is essentially one ex-
tremely long molecule, and these are freed. Like a basketball player getting
out of a Volkswagen, only more so. " He set the tube half into a fish tank full
of running water, where a thermometer said 37" C., blood heat. "Come
back in half an hour."
Out of the warm-water bath, the tube of liquid was clear again. "Now we
add phenol-what used to be called carbolic acid; our grandmothers used
it to disinfect drains. The phenol attacks the protein, which is why it worked
for grandmother, but it leaves the DNA alone. And it's heavier than water,
so it will sink to the bottom with the protein, while the nucleic acids stay in
the aqueous phase at the top." He put the stopper back into the tube, started
gently rocking it. The liquid was bubbling slightly and began to look pale
gray and thick, like sputum. "Doing this by hand keeps the DNA from
breaking so much. Those long fibres, once they're floating free, are exposed
to a lot of shearing force. That disgusting glob of white is the protein. In a
minute we'll centrifuge them apart." This time he used a small machine
standing on his bench, mushroom-shapkd, of gray metal. First he weighed
the tube, and filled a second one with which to balance the machine. A
hand-lettered sign on the centrifuge said "four buckets are hot--beware."
As he closed the lid, the spin was starting to tilt the tubes up into the
horizontal plane. "Come back in half an hour."
As we waited for the spinning to stop, Altman handed me a small brown
bottle. "This is what the stuff we're preparing would look like if you made
a lot of it and dried it. It's purified DNA, as it happens from calf thymus
glands, one of the traditional sources." The bottle was full of small bits of
what looked like scraps from an old linen handkerchief, white and ob-
viously fibrous. "They're a lot tougher than lint, though. Those long mole-
cules lying together have a very high tensile strength. No, don't touch, even
slight impurities can start breaking the molecules up." With tweezers, he
took out the top flake, dropped it into a small vial with a screw lid. "Some-
thing to show your friends."
The tube from the centrifuge now had a layer of clear liquid at the bottom,


                              G  "He was a very remarkable fellow." o 33
"the phenol," then a layer of white, "the protein-really goopy," and above
that another layer of clear liquid. "The DNA is in that top layer. It gets so
viscous we'll have trouble getting it out without contaminating it with some
of the protein beneath. If this were real, I'd have to be a lot more careful."
He packed the tube and assorted glassware into shaved ice in a battered
plastic ice bucket. He took a wide-mouthed glass pipette and tried to suck
up the layer of liquid at the top of the tube; every time he lifted the pipette
slowly away, the liquid simply plopped back into the tube. "It's so viscous it
pulls itself out again." He tried other pipettes, at last took a small one over
to a bunsen burner, where he held the tip in the flame, and then bent a kink
into it. With this he got the liquid up a few drops at a time, and into another,
narrower tube. He tilted that. "Watch how it flows. That's really viscous!
It's good stuff.
"Now, this last step is the spectacular one. I'm going to layer in twice as
much ethanol, absolute alcohol." He poured the alcohol slowly down the
side of the tube so that it floated, cream on Irish coffee. "Now we stir with
this glass rod, gently winding. The alcohol precipitates the DNA, and we
pull the fibres out of solution like winding spaghetti out of sauce." As he
twisted the glass rod, the tip began to thicken with cobwebs, wet and trans-
lucent. As he lifted the rod out, the attached fibre pulled into a long fila-
ment. "Amazing. That's a lot of individual molecules lying together, of
course. But you could use a fibre like that for X-ray analysis of its structure,
the way Rosalind Franklin did."
The principle of DNA extraction is simple: break the cells open, get rid of
protein by treatment with phenol or chloroform, precipitate DNA with al-
cohol. High-school pupils these days learn to extract DNA in science
classes, though not, if their teacher is wise, from bacteria; and teacher
scrambles to keep ahead with the aid of manuals and source books that tell
how to go on to analyze the composition of the DNA by such techniques as
chromatography or electrophoresis, in which absurdly small amounts of
biological substances can be persuaded to separate themselves for identifi-
cation and measurement as they migrate, in solution, down a sheet of filter
paper or across a hard slab of gelatine, some {ravelling faster and farther
than others because of differences in weight or solubility or electrical
charge of the individual molecules.
Methods of such discrimination and finesse-of such chemical resolving
power-were only beginning to be available when Avery published his sur-
prising paper. Yet what he accomplished in the early forties is still re-
spected as masterly. His was far from ersatz science. The paper is marked
by the probing sensitivity with which he responded to what he did not
know. The title is as arid as any in the literature: "Induction of Transfor-
mation by a Desoxyribonucleic Acid Fraction Isolated from Pneumococcus
Type III," by Oswald T. Avery, Cohn M. MacLeod, and Maclyn McCarty.
Within, the writing is excellent- supple and taut. Procedures come across
vividly. Not just a short run of experiments is reported, but years of work
and years of pondering. The argument is tough, clear, close-grained. Sci-
entific papers in our day are written to an artificial, sterilized form; Sir


                       P
34 i The Eighth Day of Creation
Peter Medawar-an immunologist who shared the Nobel Prize for physiol-
ogy or medicine in 1960-has suggested that they are so deliberately anti-
historical as to be a deception, for "They not merely conceal but actively
misrepresent the reasoning that goes into the work they describe." Avery's
great paper, though, shares with the classics of science of previous centu-
ries at least one quality now grown rare: from the first paragraphs through
to the end, one feels an original curiosity working.
Avery was by training a physician, as were his two associates in the
paper. As specialisms then went, he was not a geneticist and not exactly a
biochemist, but an immunologist and microbiologist. That is, he worked
with microorganisms; and among them microbes rather than viruses, and
among microbes the bacteria that cause pneumonia, with the long-term
hope of developing sera with which to treat acute cases.
Microorganisms can go through as many generations in half a week as
mankind has had since history began. Each generation for a bacterium is a
doubling; many a virus multiplies by the fiftyfold or the hundredfold. The
foods such creatures use are so simple that what goes into them can be
exactly known, controlled, and compared with what they make of it. Fur-
ther, it is easy to spot variations. A bacteriologist or a molecular geneticist
can routinely select from a billion or more separate cells the single individ-
ual that possesses some particular inheritable trait. These and other advan-
tages have made microorganisms the favorite subjects for many kinds of
biology for the middle third of this century, especially for geneticists,
though right now a change is taking place, back to fruit flies, mice, and
other higher animals, because one-celled creatures are too simple, inter-
nally as well as in being one-celled, for the questions molecular biologists
have come to. Yet thirty years of intensive scrutiny (as Watson points out to
students) mean that next to man himself, the world's most thoroughly
understood organism is his small companion through life, the normally
benign intestinal bacterium Escherichia coli. Determining the results of
an experiment with bacteria can be easier than one might suppose. Often
enough a microscope is not even needed: Just examine where the bugs are
growing, on broth or gelatine in one of those fragile low-sided glass dishes,
to see the colonies' size, color, texture.
By such simple and visible criteria, the world of Avery's Streptococcus
pneumoniae, called pneumococcus, is divided into the Rough and the
Smooth. S forms are virulent. They kill laboratory animals. A bacterium of
S form surrounds itself with a plump, gelatinous capsule, which it builds
not of protein but of a complex sugar, a polysaccharide. The capsule partly
protects the bacterium from the defenses of the infected animal, and so
always goes with virulence in pneumococci. R-form bacteria have lost their
ability to make capsules and so to cause infection. (R forms are now under-
stood to be mutants that fail to make the enzyme that knits together the
capsular polysaccharide.) To get them, bacteriologists grow pneumococci
in a medium made hostile to the ancestral S form. They are called R for no
microscopic reason but because, Avery wrote, "On artificial media the col-
ony surface is `rough' in contrast to the smooth, glistening surface of colo-


                "He was a very remarkable feZlow." o 35
nies of encapsulated S cells." R and S forms of pneumococci had first been
distinguished in 1923, in London, by Frederick Griffith, a physician doing
research in the Pathological Laboratory of the Ministry of Health. Variant
virulent S types had also been found, and numbered I, II, III. These differ
much less in what can be seen in a glass dish, but can be told apart with
certainty by immunological tests. Antibody reactions are among the most
exquisitely sensitive detection systems biologists possess. Tested against
serum from the blood of rabbits that had survived infection and developed
a high degree of immunity, protein from pneumococci betrays its presence,
Avery wrote, in dilutions as high as one in fifty thousand, and the capsular
sugar in dilutions of one part in six million. Avery himself had discovered
the immune reaction to capsular polysaccharide-the first evidence that
animals can make antibodies to something not a protein-also in 1923, in
the course of the work by which his laboratory established the fixed differ-
ences among the S types of pneumococci.
Then, in 1928, Griffith in London had published a startling discovery. He
had injected mice with two pneumococcal preparations at the same time:
a small amount of a living culture of the R form, derived from Type II and
proved to be not virulent by itself, together with a large amount of a dead
culture of the S form of Type III-killed by heat, containing no living bac-
teria, and proved to be not virulent by itself. In short, two different types,
one an R, was live but not virulent, and the other an S, virulent but killed.
Many of the injected mice had died. In the heart's blood of these mice,
Griffith had found living, virulent pneumococci, of the S form-and not
Type II but Type III.
The change was permanent and inherited. Generations more of culturing
had produced nothing but more Smooth, virulent Type III pneumonia
germs. Other experiments produced similar transformations of other pneu-
mococcal types. Griffith's discovery suggested doubts about the existence
of distinct true-breeding species among bacteria. It opened grave practical
problems for epidemiologists and immunologists. It raised clouds of specu-
lative and spurious explanations. All in all, microbiologists found transfor-
mation of bacteria about as unsettling as atomic physicists, at that same
time, were finding the transmutation of elements by interaction with neu-
trons and protons. Avery at first found it impossible to credit Griffith's
paper. The findings seemed to overthrow his own fundamental demonstra-
tion of the fixity of immunological types. But bacterial transformation was
confirmed that same year in Berlin and in 1929 was repeated at the Rocke-
feller Institute.
Two years after that, associates of Avery's found that they could do the
same experiment leaving out the mice. They could achieve transformation
by growing a culture of R form in a glass dish in the presence of heat-killed
pneumococci of the S form. Several months later, James Lionel Alloway,
again in Avery's laboratory, took the pursuit of the transforming agent one
twist further. Alloway broke open the S-form bacteria to set their contents
free, then passed the culture through so fine a filter that the shells, together
with any unbroken cells, were removed. When this extract, free of cells,


36 o The Eighth Day of Creation
was added to a growing culture of the R form, transformation took place.
Further, when he added alcohol to the extract, he got a viscous, "thick
syrupy precipitate."
Rollin Hotchkiss, who joined Avery's laboratory in 1935, recalled in a
biographical memoir thirty years later that Avery's characteristic question
was a quick, insistent "What is the substance responsible?" For the next
decade, Avery was increasingly preoccupied with step-by-step purification
of the transforming agent and its identification. In the beginning, transfor-
mation was an uncertain, delicately balanced phenomenon. "Many are the
times we were ready to throw the whole thing out of the window!" Avery
said to Hotchkiss. At the last, Avery was able to take a culture of pneumo-
cocci of an R form that had been attenuated from an S of Type II, thirty-six
generations back (all the way back to the Crusades, on a human time
scale), and add to it what he knew to be a highly purified DNA extracted
from an S of Type III-and he got out, in the next generation, fully devel-
oped "large, glistening, mucoid colonies" of S Type III. These then re-
mained stable through succeeding generations. Recalcitrant strains of bac-
teria had been tamed, finicky conditions of culture mastered. Avery's
difficulty was no longer the transformation itself but to prove that it was
caused by DNA and nothing else, despite the fact that DNA had not been
identified in pneumococci before and in defiance of the universal convic-
tion, his own conviction at the start, that DNA was a monotonous molecule
and genes were protein. Avery was a small man, a bachelor all his life,
smooth-faced and thin; he wore pince-nez. Various friends remember that
he would pass his hands across his bald head when perplexed, that he
rolled his own cigarettes, that he was fastidious with words and reserved
with conclusions, that he was a gentle, versatile, overwhelming monolo-
guist for whom the pneumococcus was the microcosm of biology. Hotchkiss
wrote, "My personal notes of 1936 record that in one of his discourses on
transformation, Avery outlined to me that the transforming agent could
hardly be carbohydrate, did not match very well with protein, and wistfully
suggested that it might be a nucleic acid!"
Throughout the paper of 1944, with immaculate caution, Avery, Mac-
Leod, and McCarty speak of their substance as "the transforming princi-
ple." To get it, they grew virulent Type III pneumococci at blood heat in
twenty-gallon vats of broth made from beef hearts, spun out the bacilli in
an iced centrifuge, suspended them in brine, and brought the "thick,
creamy suspension of cells" quickly to a temperature hot enough to kill the
cells and to inactivate "the intracellular enzyme known to destroy the
transforming principle" (an enzyme now called, with brisk inelegance,
DNase). They then washed the cooked pneumococci in three changes of
brine to remove capsular sugar as well as whatever protein would come
away, extracted the bacteria by shaking them for an hour in a solution of
bile salt to break the cell walls (and then threw away the cell residue), and
reprecipitated the extract with pure grain alcohol.
"The precipitate forms a fibrous mass which floats to the surface of the
alcohol and can be removed directly by lifting it out with a spatula," the


                 "He was a very remarkable fellow." o 37
paper said. This was now washed several times with chloroform to remove
protein, and suspended yet again. A digestive enzyme was put in to eat
away any remaining capsular sugar. Removal of protein was repeated,
"until no further film of protein-chloroform gel is visible at the interface."
Pure grain alcohol was added again, "dropwise to the solution with con-
stant stirring." At a concentration where the alcohol nearly equalled the
extract, "the active material separates out in the form of fibrous strands
that wind themselves around the stirring rod. This precipitate is removed
on the rod and washed. . . . The yield of fibrous material obtained by this
method varies from ten to twenty-five milligrams per seventy-five liters of
culture" -or, at best, just under one hundredth of an ounce from twenty
gallons of culture. The method of extraction, before the introduction of
detergents and using chloroform rather than phenol, was heroically labori-
ous.
Avery and his colleagues set out to show what their transforming agent
was-and what it was not, which was harder. They devised tests with an
almost obsessive ingenuity that makes the paper a model of reasoning from
and about experiment. The understated iteration takes on rhetorical power.
Standard qualitative tests for protein- for example, add a pinch of copper
sulphate and see if the solution turns blue-violet-were negative; those for
DNA, strongly positive. Chemical analysis found the elements in propor-
tions-particularly the telltale ratio of nitrogen to phosphorus, 1.67 to 1 on
average-which agreed closely with what DNA, with its nitrogenous bases
and its phosphates, should show but which would have been different if,
despite the extraction methods, much protein had remained.
They turned to enzymes. The specificity and speed of enzymes, the power
of each to catalyze its own reaction intensively and nothing else, fits them
for the burden of proving a biological negative. Pure, crystalline enzymes
were just beginning to be available, in great part through the work at a
sister unit of the Rockefeller Institute, in Princeton. Other enzymes of
proven strength in crude form were obtained from rabbit bones, swine kid-
neys, and the mucus of dogs' intestines. Of these, enzymes known to digest
proteins left the transforming principle intact. Those known to degrade
RNA left the transforming principle intact. And those that ignored protein
but attacked samples of known DNA destroyed completely the activity of
the transforming principle. Avery and his co-workers complicated the en-
zymatic tests by adding selective chemical inhibitors and by exploring the
subtle effects of temperature variations on enzyme activity. Results always
agreed, they reported, with what happened in parallel experiments with
known DNA.
They went on to immunological tests. These demonstrated that neither
pneumococcal protein nor capsular polysaccharide was present in the
transforming extract up to the extreme limit of sensitivity of the technique.
They spun a sample of the extract on the ultra-high-speed centrifuge, and
found that as it sedimented, "the material gave a single and unusually
sharp boundary indicating that the substance was homogeneous and that
the molecules were uniform in size and very asymmetric"; the result


38 o The Eighth Day of Creation

matched with DNA from calf thymus. They tried electrophoresis, and found
that as the molecules in a solution of the transforming principle were pro-
pelled by a weak electric current, they stayed together as one substance-
and that this moved relatively fast, as nucleic acids do. They found that
the transforming principle absorbed ultraviolet light at certain wave-
lengths to yield the same profile as nucleic acids. They found, and saved for
last, that the transforming principle could demonstrate its transforming
power in extraordinarily small amounts-down to "a final concentration of
the purified substance of 1 part in 600,000,000" of the culture medium
containing bacteria of R form.

Avery's concluding discussion is one of those precursors that can some-
times be looked back to in science, or for that matter in philosophy or eco-
nomic theory or painting, where one seems to see an idea struggling to
shake free from a net of previous conceptions. Strikingly,

the substance evoking the reaction and the capsular substance produced
in response to it are chemically distinct, each belonging to a wholly
different class of chemical compounds.
The inducing substance, on the basis of its chemical and physical
properties, appears to be a highly polymerized and viscous form of so-
dium desoxyribonucleate. . . . The experimental data presented in this
paper strongly suggest that nucleic acids, at least those of the desoxyri-
bose type, possess different specificities as evidenced by the selective
action of the transforming principle.

And those are the attributes of a stuff that is heterocatalytic-that can, as
the gene must do, cause the cell to make another specific substance unlike
itself. In support of that, Avery also observed,

Attempts to induce transformation in suspensions of resting cells held
under conditions inhibiting growth and multiplication have thus far
proved unsuccessful, and it seems probable that transformation occurs
only during active reproduction of the cells.

But was the transforming principle autocatalytic as well?

Once transformation has occurred, the newly acquired characteristics
are thereafter transmitted in series through innumerable transfers in
artificial media [that is, repeated generations each started in fresh broth1
without any further addition of the transforming agent. Moreover, from
the transformed cells themselves, a substance of identical activity can
be again recovered in amounts far in excess of that originally added to
induce the change. It is evident, therefore, that not only is the capsular
material reproduced in successive generations but that the primary fac-
tor, which controls the occurrence and specificity of capsular develop-
ment, is also reduplicated in the daughter cells.

Thus Avery circumnavigated the definition of the gene. He was clear and
firm about what he had demonstrated; he would not leap.

Assuming that the sodium desoxyribonucleate and the active principle
are one and the same substance, then the transformation described rep-


             "He was a very remarkable fellow." o 39
resents a change that is chemically induced and specifically directed by
a known chemical compound.

There was an irrepressible doubt-and the call to resolve it by a new order
of scientific precision.

One must still account on a chemical basis for the biological specificity
of its action.

So the conclusion checked and stumbled:

It is, of course, possible that the biological activity of the substance de-
scribed is not an inherent property of the nucleic acid but is due to mi-
nute amounts of some other substance adsorbed to it or so intimately
associated with it as to escape detection. . . . If the results of the present
study . . . are confirmed, then nucleic acids must be regarded as possess-
ing biological specificity the chemical basis of which is as yet undeter-
mined.

That far, but in public no farther. Privately, Avery did go beyond that. A
year before the results were published, he wrote a long letter to his brother,
Roy, a bacteriologist then at Vanderbilt University. The letter was medita-
tive, speculative, full of unassuming charm: it defines poignantly the sense
of responsibility to science that some acknowledge. Avery first reviewed
the years of searching, and then wrote:

Try to find in that complex mixture, the active principle!! Try to isolate
and chemically identify the particular substance that will by itself when
brought into contact with the R cell derived from Type II cause it to
elaborate Type III capsular polysaccharide, & to acquire all the aristo-
cratic distinctions of the same specific type of cells as that from which
the extract was prepared! Some job-full of headaches & heart breaks.
But at last perhaps we have it.

He described the experimental tests, and wencon:

In short, the substance is highly reactive & . . . conforms very closely to
the theoretical values of pure desoxyribose nucleic acid (thymus type)
Who could have guessed it? . . .

If we are right, & of course that's not yet proven, then it means that
nucleic acids are not merely structurally important but functionally ac-
tive substances in determining the biochemical activities and specific
characteristics of cells-& that by means of a known chemical sub-
stance it is possible to induce predictable and hereditary changes in
cells. This is something that has long been the dream of geneticists. . . .
Sounds like a virus-may be a gene. But with mechanisms I am not now
concerned-one step at a time. , . . Of course the problem bristles with
implications. . . . It touches genetics, enzyme chemistry, cell metabo-
lism & carbohydrate synthesis-etc. But today it takes a lot of well doc-
umented evidence to convince anyone that the sodium salt of desoxyri-
bose nucleic acid, protein free, could possibly be endowed with such
biologically active & specific properties & that evidence we are now
trying to get. Its lots of fun to blow bubbles,-but it's wiser to prick them
yourself before someone else tries to.


40 o The Eighth Day of Creation
Opposition to any identification of DNA as the stuff of the gene was
peculiarly concentrated at the Rockefeller Institute. Levene had been
there, active until his death in 1940, the world authority on the chemistry
of DNA and originator of the tetranucleotide hypothesis, by which repeti-
tive scheme DNA could not possibly specify diversity. Alfred E. Mirsky,
working in biochemical genetics there, was convinced and trying to prove
that the protein associated with nucleic acids in the chromosomes of higher
organisms was the active component. Mirsky argued implacably for many
years, both within the institute and in public, that some proteins are resis-
tant to the digestive enzymes used by Avery and his colleagues, so that the
DNA must have been contaminated by significant traces of active protein.
And every thought and argument there, not least in Avery's own lab, was
shadowed by memory of a cautionary triumph of a group at the institute
more than a decade earlier-the proof that enzymes are proteins and the
humiliation of Richard Willstatter. In Munich, in the early twenties, Will-
statter-who was perhaps the foremost organic chemist of the day, and a
specialist in enzymes-had claimed that he had gotten enzymatic, catalytic
action with preparations that were free of protein. On his evidence, it came
to be widely accepted that the biological specificity of solutions containing
enzymes was not due to protein. But in 1930, John Howard Northrop at the
institute crystallized pepsin and showed that it was protein. That in itself
was the second such demonstration, four years after James Sumner had
done the same with urease; but Northrop and his associates developed
precise techniques for correlating enzyme activity with the quantity of pro-
tein present, and showed conclusively that Willstatter's experiments had
been contaminated by slight traces of protein. A laboratory colleague of
Avery's for many years, Rene Dubos, when asked about the effect of the
Willstatter scandal on Avery, replied, "It was on everybody's mind!"
Avery's work, even before the paper came out, was widely though un-
evenly known, for his laboratory had many visitors. Some papers are great,
of course, because they establish, define, settle their issues. This great
paper did something else: Avery opened,a new space in biologists' minds-
a space that his conclusions, so carefully hedged, could not at once fill up.
The question was acute: If DNA is the carrier of hereditary specificity,
how? Two scientists in particular were shocked by that question into the
lines-two very different lines-their research took henceforth. Erwin
Chargaff, an established biochemist then in his late thirties, on reading
Avery's paper switched the work of his laboratory to the study of nucleic
acids. Joshua Lederberg, just graduated from Columbia University at the
age of nineteen and about to start his doctorate, found the pleasure of
reading Avery's paper "excruciating'`-so he noted at the time-and its
implications "unlimited." He decided that these implications would never
be cleared up unless bacterial inheritance could be analyzed by the meth-
ods of genetics. But bacteria were generally believed to be asexual, primi-
tive creatures incapable of exchanging genetic information. To do genetics,
Lederberg first had to show that their life cycles had a sexual stage-that
they mated. On 8 July 1945, he noted down an idea for an experiment to


"He was a very remarkable fellow." o 41

demonstrate genetic recombination in bacteria. The recruiting of Chargaff
and the mobilization of Lederberg were among the most important effects
of Avery's paper.

Avery's public caution stands in awkward contrast to the self-assurance
of Watson and Crick nine years later. The cost may have been great. The
Nobel Prize selectors had their attention drawn to Avery's work. They
waited for the second round of discoveries. Avery was sixty-seven when the
paper appeared; it was, Chargaff wrote in tribute, "the ever rarer instance
of an old man making a great scientific discovery. It had not been his first.
He was a quiet man; and it would have honored the world more, had it
honored him more." Avery died in 1955.

I asked Crick one day about boldness and caution. "Some people of course
are extremely cautious," he reflected. "Avery, exactly; he only put it in his
letter to his brother. Boldness? I would have said that Bragg and Pauling
were the people who most influenced me in these matters of style, and both
have had that characteristic. Pauling to the point of rashness. I mean, one
always knew about Linus that he would probably show an idea even if he
realized, even if he knew there was a good chance of being wrong. In fact
a lot of his ideas were wrong. But the ones that were right were important,
and therefore he was forgiven for the fact that his structure of collagen was
nonsense, for example, because the alpha helix and the pleated sheet were
fine. But what I learned from Bragg was to grasp for the essence of the
problem-and then when you've got something, get on with it and by and
large publish it reasonably quickly. Though let me tell you that in the past
I've often been dilatory, being from time to time of a lazy temperament.
But more-from Bragg and Pauling I learned how to see problems, how not
to be confused by the details, and that is a sort of boldness; and how to
make oversimple hypotheses-you have to, you see, it's the only way you
can proceed-and how to test them, and how to discard them without get-
ting too enamored of them. All that is a sort of boldness. Just as important
as having ideas is getting rid of them. And you rea_ize that in those days,
when we were working on DNA, the pace-things were very much quieter
then. Now at the moment I've been thinking a lot about this problem of the
chromosome structure of higher organisms, and the rumor of this has got
around-the pace has got so much more hectic. I must tell you I prefer the
older style-but what can one do! When we started we were living in the
woods and now here we are in the middle of a city."

b

At Harvard, in the early seventies, Watson several times gave an advanced
undergraduate course in the biology of viruses that cause tumors in ani-
mals. One September, I went to hear his opening lectures. The course was
called Biochemistry 165, and met Tuesdays and Thursdays at eleven
o'clock. Watson's office was in the solid, shabby, red brick Biological Labo-