SPECIAL SECTION: MATHEMATICS

Mathematics and biology*

Vldyanand Nanjundlah
lndWI IM!llute of St llft(l IJ!dj!~~Jta~ftl Nehnt Ctlllrr (or AdVl nefd Stllftliflt Rmmh, ll!~Ott ~o O12, India

is inappropriate m the contex1 of biology, almost certai-

From lhe Wl!IJ.lmown saying. 'All rhemlslry Is physics and
all biology Is chemistry', one might be tempted to con- nly so in the cas~ of biological entities that are at least as
clude that because pb}~ical Jaws are most compactly complex as a ceU·.
exprtSSM in the language of mathematics, biology too
can, or should, be 'matbematized'. On the other band,
In popular perception, mathematics and biology are as Reductlonlsm and deteanlnlsm
far apart as Is possible for any two fields of enquiry to
be. Mathomalks Is bellomd to be analyllcal and deduc- In what follows. by reduction I mean explaining something
tive whereas biology Is lboughl or as lar9ely descriptive In tenns something else, the 'something else' usually con-
ud ba.sed on Induction. Admittedly, this overslmpll- sisting of entities at a lower level of organization than
nes a complex Issue. Nevertheless, the fact stands out, what is sought lO be explained. An outcome is said to be
that on lhe whole lhe pndlce of blo.logy - AS reflected detennined if, given a cert.nin initial condition, it is ineVllll·
in le.chlng programs ud research pubUcatlons - Is
free of mathematical language nnd S}111bols lo an extent hie. Inevitability implies that it is the consequence of a
that would be .,.,,lbinkab~ In contemporary physics or 'law'. rule, or set of rules. A rule may be expressed in or-
chemistry. The reasons for this are many. Ao impor- dinaty language or it may be put tersely in the form of a
tant one Is that biological systems confinu. to lhe laws matbematical expression. Roughly speaking, reductionism
of physics but are not necessa11· consequences of lbe is equivalent lo the assertion that the whole is explainable
Jaws. Mathematics can illuminate aspects of lbe struc• in tenns of ils parts, and detenninism is equivalent to saying
lure or function of a llvlng system aod cu also be used that a specified initial condition leads unambiguously to a
to mab sense of the statlstlcnl consequences of dealing particular end result. A system lhat is detennined lin this
'tilb large aggregates of genes or ceUs or organisms. All sense) is often. but not always, predimble. Chaotic ~terns
tbe Sllllo, II ls dJfflcUll lo (Olltel\li of ii IMng tnllUft, are well-known exceptions: in their case, seemingly lnsigoi-
l4bn AS n whole, AS being reduclblt to lllllhemntlcal for- ftcant differen= in initiol conditions can lend lo mojor
maUsms such as those that embody lbe laws of physics.
~ dlfflculty stems from the manner In which evolution
differences in the outcome. In principle, reductionism is
bas shaped the history of life on eartb. Historical con• always V6lid. If ao object is composed of structur6llv or
tlngenc~. lhe random utnre of mntatloos and evoln- functionaUy distinct parts, the object must be fully expli-
tlonary opportunism all make IL dlfflrult to encompass cable in terms of the properties of those parts and their
biology witbln a :malhematical framework. interactions. But reduction may or may not be useful:
consider for example the case of a paioting.
IN a famous article, the physicist Eugene Wigoer expanded One can thiok of three roles for mathematics in biology.
on what he called 'The unreasonable effectiveness of The first is an extension of its role in physics or chemisliy.
mathematics in the natural sciences·. According to Wigoer. it would follow as a consequence of the reduction of biology
it was extraordimuy that mathematics - • set or fonnol rules to either of those two sciences, were such o thing possible.
for manipulating abstract symbols and • creation or the The second role is • modern one. It originotes in the ex-
human mind - could explain the messy and complicated pectation, fuelled ongiMlly by the success of Mendel's
real world 1• Others hove drawn ottention to the astonish• principles and more recently by the discovery of the genetic
ing ract that a simple algebraic equation that a schoolchild code'. that liviog creatures are essentially infonnational
cao understand, the inverse-square law of gravillltion, entities. The underlying hypothesis is that biological infor-
bolds wilbin it the key to 'the motions or the planets, the mation - the infOllMtion required for making a plant or an
comets, the moon, and the sea' (Newton. from the first animal - is encoded in terms of a set of rules, sometimes
edition of the Principia). The present essay aims to make referred to as a program. The program is said to determine
lbe point that the lean 'natural sciences' in Wigoer's title the organism. A reliable set of procedures, an algorithm, leads
one from the program to the organism. For example, the
developmenl of a multicellular orvarusm from a fertilized
egg has been compared lo an algorithmic process. The
''Tbb article Is adapted ftom 'The role ot m1UiemaUQ In bloloqy',
!Economic and Polllltal W«kly, XXXVIII (3$1. 3671-3677; Aug 30-
third role for mathematics, Including of course computa•
S.p 0$. 20031, uonol mathematics. is more convent.iollftl, Mathemotics is
••nwl· vldY•@c,,.,11.c.em..u• an aid to orgonized thought ond a bridge betwoon emp1ri•
388 CURJ\EJ,T SCIENCE, VOL. BB. 1'0 . 3, tOFEBRUARY 200~
 SPECIAL SECTION: MATHEMATICS
cal knowledge and theoretical models. Similarly, the guage. The program was believed to reside in sequences
computer is an invaluable tool when it comes to handling of ONA and the organism wa:s thought of as the output of
large masses of data or carrying out many intricate logi- the program. Salvador Luria, one of the pioneers of mole-
cal operations at great speed. The first two roles are the cular biology, has put it pithily: 'Like the computer, the
more interesting ones, and their validity depends on programmed organism bas a tape containing the instruc-
whether it is useful to think of biology in deterministic or tions (genes, ONA) and the machinery lo implement the
reductionist terms. An old-style biologist (say a biologist of instructions on the tape'6,
the 19th century or earlier) would have questioned boUJ. In short, molecular biology contains within itself both
most likely on the grounds that organisms exhibit emergent reductionist and determinist aspirations. These are, respec-
properties that cannot be reduced to the properties of their tively, that life can be explained in terms of its molecular
constituent parts. However, the biology of the lotter half constituents, and that the properties of an organism are
of the 20th century and since has been chafl>Cterized by deducible from its ONA. The reductionist hypothesis re•
an approach, known as molecular biology, in which the mains unquestioned in principle, But in practice, and this
themes of reductionism and determinism are deeply em• is true of reductionism in general, the hypothesis is not
bedded. useful beyond a particular lev,el of organization. Once one
stans looking at an entity sufficiently removed from
Molecular biology genes and proteins. SO)' the cell (let alone the organism),
new attributes may emerge a·nd entirely new explanatory
The announcement of the double helical structure of DNA concepu may be called for. To take a simple example, water
set the seal on a notion !hat was already gaining accep- is made up of hydrogen and oxygen atoms and water
tance, namely, that heredity bad a chemical basis: Mendel's 'finds its own level'. But it would be a waste of time to
particulate units of inheritance (genes) turned out to be try and explain the latter observation in terms of the former.
molecules of DNA'. Today we know that the proteins that Similarly, it is not so much that a reductionist approach to
characterize living cells are synthesized on the basis of biology is wrong. Rather, and especially if the object of
certain rules of correspondence relating sequential subsecs interest is the whole organism, it provides little insight.
of ONA (base triplets) to sequential subsets of proteins What about the second possibility? Increasingly, the pros-
(amino acids), In parallel, we have discovered that develop- pects for genetic determinism too look dubious'. In both
ment, the transformation of an egg into an adul~ is oecom- cases the reason has to do with the nature of life, and in
panied by changes in the patterns of activity of DNA. The particular, 11~th the role pla)•ed by evolution in shaping
patterns differ from one cell to another and are altered living foaos.
when the DNA undergoes heritable changes. Thus the same
chemical, ONA, is central to both heredity and developmenL The natun or life
The interesting thing about the chemistry involved in
making proteins from ONA is that it depends solely on It is notoriously difficult to glve a crisp definition of life.
interpreting the informatio11 contained in ONA; the ONA Taken together, however, living creatures exhibit features
itself remains unchanged. In consequence, a ONA sequence that stand out from those of n.on-living entities. One might
can be thought of as an encoded or symbolic represenla- summarize them by saying that being alive is a property
Uon of a particular protein. Tbls l'l!dlization unleashed a possessed by temporary, open, organized forms of matter
burst of research in biology ond set in motion on approach that can store and transmit information ond evolve by
to understanding living forms. still going strong, known natural selection. Brie0y, what this means is as follows.
as molecular biology. Fuelled bY the rate at which new focts
came lo be accumulated - on the whole, facts that were en• Living matter exists far from equilibrium
tirely unexpected, the field ecquired unprecedented prestige
and an ethos of its very own1. At ils heart, molecular biology To sustain life, organisms must constantly renew themselves.
was, and even now to some extent is. driven by an assump- Both organisms and their constituent parts can maintain
tion and a hope. The assump<ion is that the (known) rela• themselves only temporarily. Cells are re-modelled so rapidly
tionship between one ONA molecule and th.e protein it that hardly any of the molecular components of our physical
encodes offers a clue to the (unknown) relationship between body is more than a few weeks old (which makes it inter-
all the ONA molecules in an organism, the ge.nome, and the esting that we retain a contin.uous and unbroken sense of
organism itself. The hope is that this assumption can be ourselves)'.
shown to be correct - in otl)er words, that the genome can
be demonstrated lo be an implicit, symbolic represenlation Even temporary upkeep requires an input of energy
of the organism. Because both computers and molecular and infoanation
biology began to take hold of peoples' imagination at
about the same time, it became popular to compare the This process. known as metabolism, involves the chemical
genome to a computer program written in chemical Ian- transformation of the air we breathe and the food we eat.
CURRENT SCIENCE, VOL 88. NO. 3, tO FEBRUARY 2005 389
SPECIAL SECTION: MATHEMATICS
Metabolism creates new structurns in the body and repam in the genome of a species within a single generation makes
or discards old ones. Besides exchanging matter, the orga- the occurrence of any one of them at a particular time
nism continually scans the environment and extfficl.s useful unpredictable in advance. Besides, the likelihood of a muta-
information from it. Both literally and metaphorically, we tion is independent or whether it leads to a particular out-
feed on our surroundings. come or not. A mutation can be judged to be beneficial or
harmful only post facto. The judgement depends on the
Life involves organization at various levels consequence of the mutation, which in turn depends on a
in a hierarchy host of circumstances. A mutation that happens to be suc-
cessful builds on an earlier mutation that happened to be
To begin with there is molecular organization. DNA, successful Th.e complex organism that results is put to-
RNA, proteins, lipids and sugars are gigantic molecules gether via unforeseen steps and bas not come about as the
made up of anywhere from a hundred to a million atoms. result of planning. Thus, the course of evolution is ind&-
At the next level, the cells that are built from them are terminale and therefore unpredictable 12 •
orglllliz.ed into distinct compartments that subserve (in part) As if that were not bad enough, life on earth has been
different functions. Organized networks or production, affected by catastrophic accidents that disrupted and re-set
distribution and disposal coordinate metabolism. Cells form the course or natural selection. The accidents led to major
bodies made up of specialized tissues and are arranged in changes in the composition or living forms. During each
a recognizable fashion. Lastly, the reception, storage and change the relative abundance of forms was drastically
retrieval of information. both between cells and between alterl!d: a large number even went extinct. As a result, it
the body and the external world, exhibils an intricate organi- is impossible to account for what happened after any of
zation. the catastrophic episodes as a logical extension of what
was going on before. Taken together with the basic random-
Life continually evolves ness that is inherent in evolutionary change, this means that
nothing makes it necessary that a particular species of
Species change through a process whose underlying basis plant or animal should have existed in the past, or exists at
is haphazard and undirected. Despite the randomness that present (including our own species, Homo sapiens).
lies at the core of evolution, living forms give the impression To sum up, living creatures are entities in their own right
of having been designed to fit an end. Bui the impression and, as such, obey the principles of physics and chemistry.
is misleading: order results from randomness because But because the)' have an evolutionary past, they are also
randomness is filtered by evolutionary opportunism9•10. products of history. They have been shaped by a series of
The implication is that an organism is best viewed as a random steps that resulted in successful responses to con-
special-purpose device that has been shaped by a series of tingencies that were faced by their ancestors; therefore
opportunistic responses to the conditions encountered by it they are endowed with a substantial element of arbitrari-
in its evolutionary history. It is this 'special- spect ness in the way they are put together. This makes it diffi-
of organisms that leads to serious difficulties when we try cult, if not impossible, Lo think of them solely as products
to reduce biology to physics and chemistry. of physics and chemistry. Physics and chemistry set limits
within which evolution can take place but cannot specify in
Evolution and reductionism advance either its course or its end.

Opportunism is inherent in the explanation or evolution

by natural selection, first put forward by D.imin and Evolution and determlnlsm
Wallace almost JjQ years ago. Natural selection begins
with a random change, a mutation, in a DNA sequence. A cell contains encoded information within itself in the form
The change can spread through a population if, on lhe aver- of genes: every protein is the decoded version of some
age, it leads to an improvement in the individual in whom gene. Might not a living creature also be a decoded entity,
it occurred. Here 'improvement' means one thing and one the encoding being carried out by the genome as a whole?
thing only. It means that the individual is more likely to If the answer is yes, one could say that genes determine
have offspring (who inherit the changed DNA molecule) organisms. Then it would make sense to look for a theoreti-
than the average individual in the population 11. cal unde11>inning of biology in the set of rules, the genetic
What one might call a unit evolutionary episode begins program, bidden behind the organism. The programmatic
with a mutation, continues through the fate or lhe alterl!d view of development is that the genome has an algorithmic
DNA sequence over generations and, on occasion. concludes structure, that ii is put together logically, as we would put to-
with the establishment or elimination of the mutant gene. gether any set or insttuctions designed lo ensure a desired
A mutant or variant version of a gene becomes establis- end, for instance a cookbook recipe, a play or a computer
hed or 'fixed' if it replaces the pre-existing variety. The program. However, many findings point to the lack of a
enormous number of independent mutations that can occur rul&-based, logical structure to the genome as a whole1•11•
390 CURRENT SCIENCE. VOL BB. NO. 3. 10 FEBRUARY 2005
 SPECIAL SECTION: MATHEMATICS
Algorithmic operations generally lead to unique chemical reoctions, signalling in the neivous system is similar
outcomes. whereas the output or genomes to elecirical conduction through a network of wires and the
can be highly flexible now of blood depends on hydrodynamic principles. Popu-
lation genetics. a highly mathematical branch of biology,
The silllle genome, or very similar genomes. can be consisleru deals with the spread and distribution of genes in popula-
with creatures that look drastically dilferenl Sometimes this tions. Population geneticists aim (among other things) to
is because of environmental influences, sometimes because show that measurable changes in the distribution are con-
of differences in life stages and sometimes because the sistent \\1th known or assumed evolutionary forces. None
link between the genome and the organism bas an inlrinsicalJy of these cases constitute counter-examples to the point I
indeterminate component to it. Turtle and crocodile eggs have been trying to make. As an illustration of precisely
develop into males or females depending on the temperature why no~ let me take the case of something that has long
at which they are incubated; the caterpillar and the butterfly, been a popular subject among mathematically minded bio•
in •PIM'arance so different, are genetically identical: bacteria logists.
thot are members of the Sllllle clone and ore raised in the some Fifty years ago, In a publication that has been accorded
environment can differ in their obility to metabolize sugar. the status of a classic, the logician and computer scientist
Al6n Turing constructed a mathematical model for the
Conversely. very different genotypes sponlllneous origin of biological form. He began with
can be consistent with very similar organisms what he thou!l,ht was a formless structure such as a newly
fertilized egg". Twing's model was 'global': be looked al
South and North Ameriam mammals, or moDuscan and ver- the embryo as a whole, and assumed that it could be com-
tebrate eyes, are textbook examples. Thus the relationship pared to a bag \\ith chemicals that were transported by
between genomes and organisms is not one-to-one. Chim· diffusion and reacted \\ith one another. The model was
panzees and humans, different in so many respects, are said elegant and plousible. From it. Turing showed that depending
lo share 98% of their DNA sequences. 'Sibling spe on the chemicol reactions and rates of diffusion, thermo-
frogs share • much smaller fraction of ONA but require dynaniic 0uctuotions alone could leAd 10 A variety of long•
an expert to lei! opan. term outcomes. Under some conditions there was a high
concentration of chemicals at one place in the embryo
and relatively low concentrations elsewhere. Under other
The same gene or DNAsequence can contribute conditions the chemicals became distributed in spot-like
to seemingly unrelated traits or stripe-like regularities, which mimicked the pattern of
tentacles in a Hydra or arms in a starfish. Under yet other
Genes that mediate seXlllll differentiation in the fly also func- conditions the concentrations oscillated in a clocklike, and
tion in the development of their nervous system. In albino sometimes wavelike, manner.
cats, a gene that encodes an enzyme in the pathway re- Turing thought that his scheme could exemplify giobal
sponsible for the synthesis of a skin pigment, melanin, also or system-wide mechonisms for the genesis of developmenlill
affects the manner in which the eye and brain are connected. patterns. After an initially cool reception. the modei began
In shon. genomes lll'e organized very differently from LO evoke a great deal of in1eresL Later. stripe formation was
plays, recipes or computer programs. DNA sequences and shown 10 be ubiquitous in earty emb1yogenesis. For example,
their functioning renect the opporturustic fashion in wlutli even a fly embryo exhibits patterns of gene activity that
organisms have evolved. If there are rules that lie behind the resemble a zebra·s stripes; one could think that they were
ways in which plants and animals are built, the rules are Twing patterns. Unfortunate.ly, subsequent experiments
not logical consequences of the ways in which genes ore showed otherwise. Indeed. it appears that there is no global
copied or gene activity is regulated (no more, for in- rule behind stripe formation in the fly embryo. Rather, the
stance, than the functioning of human societies is a logical
manner in which the fly makes slri!M's is bizarre and con-
consequence of human biology).
founds expeciations. The stripes turn out to depend on distinct
genetic regulatory inll!ractions that iead separately to the ap•
The role of mathematics In biology pearance of each stripe. and not even in serial order at that16•
How can one account for the existence of such a non•
Undoubtedly it should be possible. and indeed is possible. to inruitive stnpe-forming system? A possible answer comes
treat a living system os one does any other physical system, from the hypoth.esis that global schemes such as the Tur-
so long os one does so within a suitably restricted framework. ing model tell us something about the evolutionary antece-
The nacrower the focus. the likelier it is that a physical or dents of presenHlay patterns - antecedents dating from a
chemical, and therefore mathematical, approach -.ill be !M'riod when evolutionary embellishments were minimal
useful 11• Many aspects of the functioning of cells and tissues and the link between genes and development was not as in-
can be subject to mathematical treatment quite successfully. timate as it is todayn. According to this proposal, global sys-
Metabolic networks can be modelled as linked cascades of tems for patterning, based on physics and chemistry, may
CURRENT SCIE~CE, VOL. 88. NO. J, to FEBRUARY ZOOS J9 1
SPECIAL SECTION: MATHEMATICS
have existed in the past. lnl!Vitably, their outcomes would sius Oobzhansky put it in the fonn of a maxim: 'Nothing
have been 'noisy', meaning subject to IMge variations, and so in biology makes sense except in the light of evolution',
unreliable. The gene-based patterning mecbdnisms lh4t we Coming back to \\Tigner, mathematics is unlikely to be 'effec-
see today could have come about in the course of evolu- tive· in biol'/llf, lei alone 'unreasonably effective', because
tion bocause they buffered the variations and ensured that of avolution 1 •
the patterns were produced reliab1yts_
Stripe fonnation illustrates ii case of ii theory not standing
up. More glaring is the absence of any theory. To realize
Notes and references
this, one has only to see that whereas a series of findings I. 1\-,gner, E. P.. Commwtications in Pure and Applied Mathematics,
have extended our knowledge of how genes. cells and Ol!Jdll- N...- Yon: John \\ru,y & Soos, Joe, 1960, vol. I3. One mlght
isms function, the bulk of the new kno\\1edge bas been in pii."lpbrase w·JgDer by eying lhat a gime imtenled b)' human beings
the form of one surprise after another. Consider some strik• for Uteir pleasure and 1mu.semeol h1ppens. just by chance, lo ex-
plain fNtms of UJe ml.I world. Wlq:ier ac.knowled11es later lhat he
ing examples: the discovery that not aU ONA codes for ts rtfemr;o lo lhe pbyslcll sciences llone f Thc ml111cle or the ep-
proteins; the fact that genes can consist of discontinuous proprt.te.na1 of lbe l111ou1oc of mltht!mnllcs for Urn formuloOcm
segments of ONA; the observation that when mulbted. of !he ltws of pbf1ICS .. .'I.but lhlS qullfication lends lo be lgoored.
some genes seem to leave the organism unchanged; and tbe In Con,'!rs,nioni on )hnd. Malitr, and MnthemaUcs (edited and
absence of any correlation between genome size and gene tnnsl!ltd by M. 8. 0.80\'0ise: Prio:olon Unlversl~• Pross, 19981,
Jta.n~Pie:rre Cbanotu1 n.nd Ala.in Co,nes ugue whttbcr matlltm11t•
number. Testifying to the absence of a theory of living 1c:s u I creatioo of tht mind or reflects the extemaJ world. It is
systems, let alone a logical theory, none of these major \\"Orth noting that biology has influenced ma.lhemaUcs in lhe past,
findings were anticipated. and lhe innueoce is likely to gfl sroog<r in lbe future. But lbe
preseot irticle does oot deal "itll tlili issue.
2. A secoDd aim is to question, if only indirectly, lbe implicit assump·
Summing up tion tbJ! • product of lbe mind can be analys,d as if it need nol
cony an imprint of the forces that sheped tbal mind. From the
Unlike physical objects, whkh can be accounb!d for as the ,iewpoiil1 of n·olotionaiy biology, a lbeol)' of knowledge must be
necessary consequences of the operation of natural laws, embedded in the e•..-olu.tionary bis10I)' of the mind. The ethologist
living entities are products or an essentially ad hoc process Konrad Lorenz 9°'" the c:Jearesl explanation or e-volullonary epis·
known as evolution. Thoy t.avo beon moulded by natural temo!ogy. Seo Oonald T. C.mpll<U. E,·olutionary Eplslomotogy In
Tb• phllo,op~y ol KAr1 R. Popper. t<!lted by P A. Schlipp, 1974,
selection in a manner that bas preserved a succession of pp 412~6J, t.aS.Ue, JL. Open Coun,
minor, randomly caused changes th.It rumed oul to be suc- l . WU?! some exc:epuons that we un lgnoni, 11 gene ls something lhnt
cessful. The properties that they exhibit demand e>-olutionary lonnally ts defined b)' the totlowlno pruperttes: II ts , mohs:ulor
explanations. An answer to lie question of jus1 what con.stl• polymer bdongu10 to Ult gt1:nerll class known OS DNA; it Is round
lutes an evolutionary explanation falls outside the scope of uwdt the nudeus of living cells; 11 can be passed in toto from a
this article, but there are at least two respects in which it
°'
fa!her mothtr to a e:bild: a.nd u Clm thanoe, ln whtth cue some
tr.its are also tbanged tll,on,by m,king il appear lbat lrAlts are
is very different from what would be called an expl4na- pi1'ed down from parents lo offspring whereas octuAlly only lhe
tion in physics or chemistry. First, in evolution the envi- QflleS are). Tbe lmk be:twten a gere an.d a trait ls via. a protein.
ronmenl, including the ancestral environment, pl4ys a cmlral The units that make up a DNA pol}10er, taktn In sequence, spec·
role in defining the organism. Second, the basic uni! of change ify - via certaiD rules of correspondence - the sequence of units
1bit mm up a protein polymer. The rules a.re known as lhe qe·
in evolution is not the individual organism. Instead, it is a oe.tic code. Taken together, the set or DNA molecules inside. a cell
collection of entities known as the species. Because of constitute the genome. Because not all DNA molecules encode
evolution, living creatures are products of history. They proleins. the genome bas botll gene.a; and non-coding DNA For a
make sense, that is, are capable of being understood, onl)' nuanced discussioo of the philosophical complexilies lurking be·
in the context of their history. Evolutionary explanations hind lhese dumltions, see GeneHc, and Reductionism by Saholra
Sartar lC4mbridge Uoivmlty Press 1998), to ANew Btolog)' lor
would be out of place in the case of purely physical ob- a Sew Century IM>trobioL Mol. Biol. Rfv.. 2004, 61, 17J- 186),
jects, For example, no one would think of saying that to C. R. Woese distingulshes between 'empirical reducllonlsm', a
understand the hydrogen atom, all the hydrogen atoms in m<lllodolog!cal tool, aod 'luncl!mettatlsl n:ducllonlsm', which 1,
the universe had lo be studiEd, in addition to how they got ex,mpllf\ed by the reducttontsm of 19th c,ntury claulcal phy1lcs,
~ J D. Wat>on 91\'e, • racy account or lbe dlsoov.ry In The Doubt,
that way. Of course observ,tion, description, experimen•
Htltt JAlll<naeum. New Yori(. 1968) Tb• Path to 11,, Double Helix
14tion, logical analysis and the construction of testable hy- br R. Otby (Un1,-mit1· or Wa.sbu,gton Prm, Seatu,, 1974) Is Ibo
potheses - what is sometimes called the method of science - m«n sc.hal6.th• hi.story. Curiously, eVtn lhouob molttular blolo<U'
are as much a part of biology as of the other natural sci- b<nefiltd enom,ously b)' the enlly cf physicists, and was bulll on
ences. But there is something that makes biology special, tilt el])<rtmeclil toots dtvised by )i>ysicislS, physics as such bas
and that is the history of change undergone by living mat- plared only a minor role in molerulu biology (a prominent excep·
tion being lhe role of X-ray CJ)'slill,grapby). Ralber, what pbysi·
ter and the manner in which the change has come abouL It cists contributed to the field was c 'physics way· of looking at
is because of this that one m11St doubt whether biology can problems. of gfiling dov.-n to essutials. Initially the}' we£? in-
ever have a mathematical structure in the way that theo- spJied by lbe hope rAlsed by Elwin Scbroding<r !hat 'we mus! b<
retical physics does. The evolutionary biologist Theodo- pr<pared lo llDd a new ~'Pf of physlcill law prevailing in ii (living

J92 CURRENT SCIENCE, VOL 88, NO, J, 10 FEBRUARY 2005

 SPECIAL SECTION: MATHEMATICS
mol!trl' (What I• Ufe7, Cillll>ridge Untvenily l'lffl, 1944 1, but tbe tradition ombli1bed by D' Arey Thompson. Wolfram ts expli<·
1h11 bop, or•dunlli· fad<d ,--.y. ltly antl·stlecUonlit. 'Wbat I bll\'C come to believe ts that tn-tn)' or
5. Mora119,, M , A litsto11· of Molecular Bloloqi·. Huvanl Unh·orslli• the most ob\'lous examples or complcx:lt)' In blologlenl J)'S1.erm
Pttss, ClmbMdoo. 2000. 4Cluall)' have \"try ldUe to do wllh adap1•Uo11 or natural sclecUocf
o. Stt paq, 4 or 30 Lec1ur<S in 81ology bys. L I.urn !MIT Pross, Apart from ,~moo m how to go about crHting an lntercsUno but
Cambridge, 19n). 0~ Th< Secrel ol Lile by J. D. Watson and 1.mao1nar)' wor1d, tbe rtlt\'ance of his auloriuta•based approach fo:
/1. W. Deny (Knopf, 20031 >dopl5 • simllar ,·1, wpoinl A rorcoful understanding biological fonn Is unclear_
m
n,butlil coo be round in Unravelling the Secrtl Life by Bany 14. This is true of ficlds ranging rrom literary studies to sports. There
COIIIJnOllfr (GeneWatcb. 2003, II, 1-8). is or course a pure.ty utilitarian role tbal mathematical or com·
7. Nijlloul, tt. F., Mell!pbors and the Role or Genes in Developmenl puter-based approte.hes ser\'e. No one questions tbe usefulness of
Biousays, 1990, 12, 44 1-146. The pbrne 'pbeno(ypic plasticity' statistical analysis whenever quantitative mtasurements are car·
expresses the observation tbct genotypes and phenotypes do not ried oul But vte are asking whether zi m U1emati..._;iJ or algon1hmic
bn-e a one-to-one relationship. Morpbogeoesis and Pattern For· sttucture might aplain biology in the way lhat ~ seems to expliin
matioo in Biological S)'Stems (eds Sekimura. T. d al.), Springer Ver· the physical uni\'ene.
lag, Tokyo, 2003 and Origination of Otyanismal Form (eds 15 Turtng, A. M.• In Tll, Chemical Billis of Morphogenesis. Philos.
Miillff, G. B. and N.,.1llall, 5. A.I, MIT - •· 2003 contain more Trans. Roy. Soc. London. 19;2, 852. 1!-152. For a dlS<us,ioo of
Oil the same theme. VtsUal pathways in albinos by R. W. Guillery tbe history of Tun1141 models >ee Alan Tunng and tho Chemic;&)
(Sd Am., 1914, 234, 44-541 contain> a strila1141 example of the Ba>ls of Morphogen..is. NanJundloh, V.• In Morphoponest, aod
compi.x rotaUonshlp belweeo g,n,. and trolb. Abo,., Marks, J , Pol!tm Fonnollon in 81oloolcal S)•stem, (eds Sektmuru, T. et al ).
ll'llet 11 monn, to bo 98~ Chlmponzct, Unh·enlly or c,11romta Springer Verl19, Tolc)'0, lOOJ, pp, 33-4),
Pm!, 2002. 16. Akam, M., Maktog ll!ipo, lneloganll)'. San,re, 1989, JU, 282-
8. One. ur, depart,, decoy uo be extroonlln•rih• rapid. II hos boon 183. Th• po1n1 is 001 chat • T11ring mt<booism is lmpomble: nor
pouuf'd out lhal after le )"tffl a discarded automobllt can JLUI be ;, It that no such tne<l!oobm exists in 11!1)' o,gaoum. The point 15
fotll!d wbert U was ltft: an a.nimal that dies m I fort!it vanishes in that the sn.me outcomt ca.n be reached by wha:t ts presumably IJl
a matter or months if oot weeb or days, ad-hoc evolutionary process.
9. The contention or e:volationa?Y theory is lhat inmimale nature and 17. See Newman, S. A.. ls segmenlation generic?. BioEssa)'S, 1993,
blind chance are saffictau to engineer oulcomes that setm to be 15, 277-283. Multirellular development. including pattern forma-
designed with a prior pwpose (Dawkins, R., ThE Blind \\iatcb- tion,. has to do with the behaviour of groll.pS of cells. It demands,
mal:er, Penguin, London, 198'.!J. therefore, a "sociologica.1' approach. The same point has been
10. Excellent accounts or t!\'Olutionary opportunism can be round in made with reganS to a.bemnt de\ielopmea1 in The Society of Cells:
The meaning of e.-olutioo by G. G. Simpson (Yile Untversi~• Cancer and control o! cell proliferation (C. Sonnenscbein and A.
Pm>. New Haven, 19671 and The Possible and lb, Actual by F. M. Soto, Sprinoor·Verlilg, New York. 19991
JKob (Untv, rslly or wa,llington Press, s,,111,.
1982). 18. C. H. Waddington mode the concept of rellaillllty or bulferiog th•
11. N1tunl selection ts the ouly Inanimate ag-ency that we know of dertntno elemen1 bb wa)· of looklng 1t den.lopment and ooln.ed
tb.tl b capable or mhnk tirt9 conscious deslgn, But all cvohltion· tbe word •can4Jb,4Uon· to describe It.
a,y cbanoc need not be because or natural seledJon. Seo ror ex• 19 Tim calls into question the tittltude lhat H trea of 11tudy can bt
amplt Ever Since Da.-10 by S. J, Gould (W. W. Norton, N,w called a sch:net only lf It supports a matlltm1tkal f'rumuwork. ~
York. 19771. malhemallcs•ln·biolo;y story ha! man)· otbtr an9les, bmong them
12. Thereby rosembhno. ft.I.bell (or entirely diffemu reasons, what sociological ones, t.Mt are not considered h«e. The tnteUedual
happens in systems Um ahtbil chaos. A nuthffllabcal N{UaUon prtstloe aMociAttd with mathematics, and l1 ottt step removed., of
m1y bt detenninistlc in tbt sense thal Lhe outcome i3 exacll)' pr~ tbtoretlcal phys.its. ha.s sometimes tended to overawe biologists.
dictable once the starting situation Is precisely spectlied. But if in· lo The Gro•1h ol Biological Thought (Hu va,d University Press,
fi!litesimat variations in initial conditions Jeod to large variations 1985) Ernst Mayr o!Im an acid co,runa,t on this tendency: 'Phys•
in the outcome, for all practical purposes the outcome is unpre- ics envy is lbe cune of biology'. /u to imman and other reasons
dictable. v1by mathemeticaJ biology bas not been '\l'flcomed by many biolo-
13. For c c'0ntrary view of pattml formation in biology, see A New gists, s,e Evelyn Fox Kfller in Mal<ing Sense of Lile IHanWII
Kind of Science (Wolfram. s .. 2002, Wolfram )fedia). following Uoivmity Press), 2002.