TrHE

AM ERICAN NATURALIST
VOL. XXX. June,I896. 354

A NEW FACTOR IN EVOLUTION.

BY J. MARK BALDWIN.

In several recent publications I have developed, fromdiffer-

ent points of view, some considerations which tend to bring
out a certain influence at work in organic evolution which I
venture to call " a new factor." I give below a list of refer-
ences' to these publications and shall referto them by number
as this paper proceeds. The object of the present paper is to
PReferences:
(1). Imitation: a Chapterin theNatural Historyof Consciousness,Mind (Lon-
don), Jan., 1894. Citationsfromearlier papers will be found in this article and
in the next reference.
(2). Mental Development in the Child and theRace (1st. ed., April, 1895; 2nd.
ed., Oct., 1895; Macmillan & Co. The present.paper expands an additional
chapter (Chap. XVII) added in the German and French editions and to be
incorporatedin the thirdEnglish edition.
(3). Consciousnessand Evolution, Science, N. Y., August,23, 1895; reprinted
printedin the AMERICAN NATURALIST, April, 1896.
(4). Heredityand Instinct(I), Science, March 20, 1896. Discussion beforeN.
Y. Acad. of Sci., Jan. 31, 1896.
(5). Heredityand Instinct(II), Science,April 10, 1896.
(6). Physical and Social Heredity,Amer. Naturalist,May, 1896.
(7). Consciousnessand Evolution, Psychol. Review, May, 1896. Discussion
beforeAmer. Psychol. Association,Dec. 28, 1895.
31
442 The AmericanNaturalist. [June,

gatherintoone sketch.an outlineofthe view of the processof

developmentwhich thesedifferent publicationshave hinged
upon.
The problemsinvolvedin a theoryof organicdevelopment
maybe gatheredup underthreegreatheads: Ontogeny,Phy-
logeny,Heredity. The general consideration, the " factor"
whichI proposeto bringout,is operativein the firstinstance,
in thefieldof Ontogeny;I shall consequentlyspeak firstofthe
problemof Ontogeny,thenof thatof Phylogeny,in so faras
the topicdealt withmakesit necessary,thenof thatof Her-
edity,under the same limitation,and finally,give somedefi-
nitionsand conclusions.
I.
Ontogeny:"Organic Selection" (see ref.2, chap. vii).-The
seriesoffactswhichinvestigation in thisfieldhas to deal with
are those of theindividual creature'sdevelopment;and two
sortsof factsmay be distinguishedfromthe point of view of
the functionswhichan organismperformsin thecourseof his life
history. Thereis, in the firstplace, the developmentof his
heredityimpulse,the unfoldingof his heredityin the forms
and functionswhichcharacterizehis kind,togetherwiththe
congenitalvariationswhich characterizetheparticularindi-
ual-the phylogeneticvariations,whichare constitutionalto
him; and thereis, in the secondplace,the seriesoffunctions,
acts, etc., which he learns to do himselfin thecourse of his life.
All of these latter, the special modificationswhich an organism
duringitsontogeny,
lundergoes throwntogether,have been called
"acquired characters,"and we may use that expressionor
adopt one recentlysuggestedby Osborn,2" ontogenicvaria-
tions" (except that I should preferthe form" ontogenetic
variations"), if the wordvariationsseemsappropriateat all.
2 Reportedin Science, April 3rd.; also used by him beforeN. Y. Acad. of Sci.,
April 13th. There is some confusionbetweenthe two terminations" genic" and
"genetic." I think the proper distinctionis that which reserves the former,
genic," forapplicationin cases in which the word to which itis affixedqualifies
a termused actively,, while the other, "genetic" conveyssimilarlyapasszwe sig-
nification;thus agencies, causes, influences,etc., and "ontogenic phylogenic,
consequences,etc, and " ontogenetic,phylogenetic,etc."
etc.," while effects,
1896.] A NewFactorin Evolution. 443

Assumingthatthereare such new or-modifiedfunctions, in

the firstinstance,and such "acquired characters," arising by
the law of " use and disuse" fromthesenewfunctions, our far-
therquestionis about them. And the questionis this: How
does an organismcometo be modifiedduringits lifehistory?
In answerto thisquestionwe findthatthereare threedif-
ferentsorts of ontogenicagencies which should be distin-
guished-each of whichworksto produceontogeneticmodifi-
cations,adaptations,orvariations.These are: first, thephysi-
cal agencies and influencesin the environmentwhichwork
upon the organismto produce modifications of its formand
functions. They include all chemical agents,strains,con-
tacts,hindrancesto growth,temperature changes,etc. As far
as these forceswork changes in the organism,the changes
may be consideredlargely" fortuitous " or accidental. Con-
sideringthe forceswhichproducethemI proposeto call them
"physico-genetic." Spencer'stheoryof ontogeneticdevelop-
ment restslargelyupon the occurrenceof lucky movements
broughtout by such accidentalinfluences. Second,thereis a
class ofmodifications whicharisefromthe spontaneousactiv-
ities of the organismitselfin the carryingout of its normal
congenitalfunctions. These variationsand adaptationsare
seen in a remarkableway in plants,in unicellularcreatures,
in veryyoung children. Thereseems to be a readinessand
capacityonthe partoftheorganismto "rise to the occasion,"
as it were,and make gain out ofthe circumstances of its life.
The factshave been put in evidence(forplants) Henslow,
by
Pfeffer, Sachs; (for micro-organisms) by Binet, Bunge; (in
human pathology)by'Bernheim, Janet; (in children) by
Baldwin (ref.2, chap. vi.) (See citationsin ref.2, chap.ix, and
in Orr,Theory ofDevelopment, chap.iv). These changesI pro-
poseto call " neuro-genetic,"layingemphasison whatis called
by Romanes,Morganand others, the" selectiveproperty " ofthe
nervoussystem,and of life generally. Third, there is the
greatseriesofadaptationssecuredby consciousagency,which
we may throwtogetheras " psycho-genetic."The processes
involved here are all classed broadly under the term"in-
i. e;, imitation,gregariousinfluences,maternalin-
telligent,"
-444 The American
Naturalist. [June,

struction,the lessons of pleasure and pain, and of experience

generally, and reasoning frommeans to ends, etc.
We reach, therefore,the followingscheme:

Ontogenetic
Modifications. OntogenicAgencies.
1. Physico-genetic.......... . 1. Mechanical.
2. Neuro-genetic............ 2. Nervous.
3. Psycho-genetic............ 3. Intelligent.
Imitation.
Pleasure and pain.
Reasoning.
Now it is evident that there are two very distinct questions
which come up as soon as we admit modificationsof function
and of structurein ontogeneticdevelopment: first,there is the
question as to how these modifications can come to be adap-
tive in the life of the individual creature. Or in otherwords:
What is the method of the individual's growthand adapta-
tion as shown in the well known law of " use and disuse?"
Looked at functionally,we see that the organism manages
somehow to accommodate itselfto conditions which are favor-
able, to repeat movementswhich are adaptive, and so to grow
by the principle of use. This involves some sort of selection,
fromthe actual ontogeneticvariations,of certainones-certain
functions,etc. Certain otherpossible and actual functionsand
structuresdecay fromdisuse. Whatever the method of doing
this may be, we may simply, at this point, claim the law of
use and disuse, as applicable in ontogeneticdevelopment,and
apply the phrase, "Organic Selection," to the organism's be-
havior in acquiring new modes or modifications of adaptive
functionwith its influenceof structure. The question of the
method of " Organic Selection " is taken up below (IV); here,
I may repeat, we simply assume what every one admits in
some form,that such adaptations of function-" accommoda-
tions" the psychologist calls them,the processes of learning
new movements,etc.-do occur. We then reach another ques-
tion, second; what place these adaptations have in the gen-
eral theoryof development.
Effectsof Organic Selection.-First, we may note the results
of this principle in the creature's own private life.
1896.] A NewFactorin Evolution. 445

1. By securingadaptations,accommodations, in special circum--

stancesthecreatureis keptalive,(ref.2, lst ed., pp. 172 ff.). This
is true in all the three spheres of ontogeneticvariation distin-
guished in the table above. The creatures which can stand
the "stormand stress"of thephysical influencesofthe environ-
ment, and of the changes which occur in the environment,by
undergoingmodificationsof theircongenitalfunctionsor of the
structures which they get congenitally-thesecreatureswill live;
whilethosewhich cannot,will not. In the sphere of neurogen-
etic variations we find a superb series of adaptations by
lower as well as higher organisms during the course of onto-
genetic development (ref. 2, chap. 'ix). And in the highest
sphere,that of intelligence (including the phenomena of con-
sciousness of all kinds, experience of pleasure an d pain, imita-
tion, etc.), we find individual accommodations on the tremen-
dous scale which culminates in the skilful performances of
human volition, invention,etc. The progress of the child in
all the learning processes which lead him on to be a man, just
illustrates this higher formof ontogeneticadaptation (ref.2,
chap. x-xiii).
All these instances are associated in the higher organisms,
and all of them unite to keepthecreaturealive.
2. By this means those congenitalor phylogeneticvariations
are keptin existence, which lend themselvesto intelligent,
imitative,
adaptive,and mechanicalmodificationduring the lifetimeof the
creatureswhichhavethem. Other congenital variations are not
thus kept in existence. So there arises a more or less wide-
spread series of determinate variationsin eachgeneration'sonto-
qenesis(ref.3, 4, 5).3
"It is necessaryto considerfurtherhow certainreactionsof one single organ-
ism can be selectedso as to adapt the organismbetterand give it a life history.
Let us at the outsetcall this process " Organic Selection" in contrastwith the
Natural Selectionofwholeorganisms. . . . If this (naturalselection)worked
alone, everychange in the environmentwould weed out all life except those or-
ganisms,which by accidentalvariationreacted already in the way demanded by
the changed conditions-in everycase new organismsshowingvariations,not,in
any case, new elementsof life-history in the old organisms. In orderto the latter
we would have to conceive . . some modification ofthe old reactionsin an or-
ganismthroughthe influence of new conditions. . . . We are, accordingly,
left to the view that the new stimulationsbroughtbychanges in the environment
446 The Americasn
Naturalist. [June,

The furtherapplicationsof theprinciplelead. us overinto

the fieldofoursecondquestion,i. e., phylogeny.
II.

Phylogeny: Physical Heredity.-The question of phylogen-

eticdevelopmentconsideredapart,in so far'as may be, from
that of heredity,is the.questionas to whatthe factorsreally
are whichshowthemselvesin evolutionary progressfromgen-
erationto generation. The mostimportantseriesof factsre-
centlybroughtto light are thosewhich show what is called
" determinatevariation" from one generationto another.
This has been insistedon by thepaleontologists.Of the two
currenttheoriesof heredity,only one, Neo-Lamnarkism-by
means of itsprincipleof the inheritanceof acquired charac-
ters-has been able to account forthis fact of determinate
phylogenetic change. Weismann admits the inadequacy of
the principleof naturalselection,as operative'onrival organ-
isms, to explain variationswhen theyare wanted or, as he
puts it, "the right variationsin the right place" (Mionist,
Jan.,'96).
I have argued, however,in detail that the assumptionof
determinatevariationsof functionin ontogenesis, underthe
principleof neurogenetic and psychogeneticadaptation,does
away with the need of appealing to the Lamarkian factor.
In the case i. g.,of instincts," if we do notassume conscious-
ness,thennaturalselectionis inadequate; butifwe do assume
consciousness, then the inheritanceof'acquired charactersis
unnecessary " (ref.5).
"The intelligencewhichis appealed to,to take theplace of
instinctand to give riseto it,usesjust thesepartialvariations
whichtendin the directionoftheinstinct;so theintelligence
supplements such partial co-ordinations, makes them func-
tional, and so keepsthecreaturealive. In the phrase of Prof.
themselvesmodifythe reactionsof an organism. . . . The facts show that
individual organismsdo acquire newadaptationsin their lifetime,and thatis our
firstproblem. If in solvingit we finda principle whichmayalso serve as a prin-
ciple of race-development, thenwe maypossibly use it againstthe ' all sufficiency
of natural selection' or in its support" (ref. 2, 1st. ed., pp. 175-6.)
1896.] A NewFactorin Evolution. 447

Lloyd Morgan, this prevents the ' incidence of natural selec-

tion.' So the supposition that intelligence is operative turns
out to be just the supposition which makes use-inheritance
unnecessary. Thus kept alive, the species has all the time
necessary to perfectthe variations required by a complete in-
stinct. And when we bear in mind that the variation re-
quired is not on the muscular side to any great extent, but in
the central brain connections, and is a slight variation for
functional purposes at the best, the hypothesisof use-inherit-
ance becomes not only unnecessary, but to my mind quite
superfluous" (ref. 4, p. 439). And for adaptations generally,
"the most plastic individuals will be preserved to do the
advantageous things for which their variations show them
to be the most fit, and the next generation will show an
emphasis of just this direction in its variations" (ref. 3, p.
221).
We get, therefore,fromOrganic Selection, certain results in
the sphere of phylogeny:
1. This principlesecuresby survivalcertainlines of determinate
phylogenetic variation in thedirectionsof thedeterminateontogen-
eticadaptationsof theearliergeneration. The variations which
were utilized forontogeneticadaptation in the earlier genera-
tion, being thus kept in existence, are utilized more widely
in the subsequent generation (ref.3, 4). "Congenital varia-
tions, on the one hand, are kept alive and made effectiveby
their use foradaptations in the life of the individual; and, on
the other hand, adaptations become congenital by further
progressand refinementof variation in the same lines of func-
tion as those which their acquisition by the individual called
into play. But there is no need in either case to assume the
Lamarkian factor" (ref.3). And in cases of conscious adap-
tation: "WWe reach a point of view which gives to organic evo-
lution a sort of intelligent direction afterall; forof all the
variations tending in the direction of an adaptation, but inad-
equate to its complete performance,onlythosewill be supple-
mentedand keptalive whichtheintelligence ratifiesand uses. The
principle of 'selective value' applies to the othersor to some of
them. So natural selection kills offthe others; and thefuture,
448 Th7e Naturalist.
American [June,

developmentat each stageof a species' development

mustbe in the
-thusratifiedby intelligence.So also with iinitation.
directions
Onlythoseimitativeactionsof a creaturewhichare usefulto
him will survivein the species, forin so faras he imitates
actionswhich are injurioushe will aid natural selectionin
killinghimselfoff. So intelligence,and the imitationwhich
copiesit,will set the directionofthedevelopmentof the com-
plex instinctseven on the Neo-Darwiniantheory;and in this
sensewe may say thatconsciousness is a ' factor'" (ref.4).
2. The mean of phylogenetic variationbeingthusmade morede-
terminate,furtherphylogeneticvariationsfollow aboutthismean,
and thesevariationsare again utilizedby OrganicSelectionfor on-
adaptation. So thereis continualphylogeneticprog-
togenetic
ress in the directionsset by ontogeneticadaptation (ref.3, 4,
5). "The intelligencesupplementsslightco-adaptationsand
so givesthemselectivevalue; but it does notkeep themfrom
gettingfarther selectivevalue as instincts, etc.,byfar-
reflexes,
ther variation" (ref.5). "The imitativefunction,by using
muscular co-ordinations, supplementsthem,securesadapta-
tions, keeps the creaturealive, preventsthe 'incidence of
naturalselection,' and so givesthespeciesall thetimenecessary
to get the variationsrequiredforthe fullinstinctive perform-
ance ofthefunction " (ref.4). But," Consciousimitation,while
it preventsthe incidenceof naturalselection,as has beenseen,
and so keepsalive thecreatureswhichhave no instincts forthe
performance oftheactionsrequired,nevertheless doesnotsub-
servethe utilitieswhich the special instinctsdo, nor prevent
them fromhaving the selectivevalue of which Romanes
speaks. Accordinglyon the moregeneraldefinition of intel-
ligence,which includes in it all consciousimitation,use of
maternalinstruction, and thatsortof thing-no less than on
the morespecial definition-westillfindthe principalof nat-
ural selectionoperative" (ref.5).
3. This completelydisposes of the Lamarican factor as far as
two lines of evidencefor it are concerned. First,the evidence
drawn fromfunction, "use and disuse,"is discredited;since
by " organic selection,'"the reappearance,in subsequent
ofthevariationsfirstsecuredin ontogenesisis ac-
generations,
1'S96.] A NewFactorin Evolution. 449

counted for withoutthe inheritanceof acquired characters.

So also the evidence drawn frompaleontologywhich cites
progressivevariationsrestingon functionaluse and disuse.
Second, the evidence drawn fromthe factsof "determinate
variations;,"sinceby thisprinciplewe have the preservation
ofsuch variationsin phylogenywithoutthe inheritanceofac-
quired characters.
in theold sense; since the adpa-
4. But thisis not Preformism
tationsmade in ontogeneticdevelopmentwhich "set" the direc-
tion of evolutionare novelties of function in whole or part
(although they utilize congenital.variations of structure).
And it is onlybytheexerciseofthesenovelfunctions thatthe
creaturesare keptalive topropagateand thusproducefurther
variationsof structurewhich may in time make.the whole
function,withits adequate structure, congenital. Romanes'
argumentfrom" partialco-adaptations " and " selectivevalue,"
seemto hold in thecase ofreflexand instinctive functions (ref.
4, 5), as against the old preformist or Weismannistview,al-
thoughthe operationof OrganicSelection,as now explained,
rendersthemineffective whenurged in supportof Lamark-
ism. "We may imagine creatures,whose hands were used
forholdingonlywiththethumband fingerson the same side
ofthe object held,to have firstdiscovered, understressof cir-
cumstancesand withvariationswhich permittedthe further
adaptation,how to make use of the thumbforgrasping op-
positeto the fingers,as we now do. Then let us supposethat
thisprovedof such utilitythat all theyoung thatdid not do
it were killed off; the next,generationfollowingwould be
plastic,intelligent,or imitative,enough to do it also. They
woulduse the same co-ordinations and preventnaturalselec-
tion getting its operation on them; and so instinctive
'thumb-grasping' mightbe waited forindefinitely by the
species and thenbe got as an instinctaltogetherapart from
use-inheritance " (ref.4). " I have cited'thumb-grasping'be-
cause we can see in the child the anticipation,by intelligence
and imitation,of the use of the thumb forthe adaptation
which the Simian probablygets entirelyby instinct,and
whichI thinkan isolatedand weak-minded child,say,would
also cometo do by instinct"' (ref.4).
450 TheAmericanNaturalist. [June,

5. It seemsto me also- thoughI hardlydare ventureintoa

field belonging so strictlyto the technicalbiologist-that
this principle mightnot only explain many cases of widespread
"determinatevariations" appearing suddenly,letus say, in fossil
deposits,but thefact that variations seem oftento be " discon-
tinuo'us." Suppose, for example, certain animals, varying,
in respectto a certain quality,froma to n about a mean
x. The mean x wouldbe the case mostlikelyto be preserved
in fossilform(seeing that there are vastly more of them).
Now supposea sweepingchangein the environment, in such
a way that only the variationslying near the extremen.
can accommodateto it and live to reproduce. The next
generationwould then show variationsabout the mean n.
And thechancesoffossilsfromthisgeneration, and thesubse-
quent ones, would be of creaturesapproximatingn. Here
would be a great discontinuity in the chain and also a wide-
spreadprevalenceofthesevariationsin a setdirection. This
seemsespeciallyevidentwhenwe considerthatthe paleontol-
ogist does not deal with successive generations,but with
widelyremoteperiods,and thesmallestlapse oftimewhichhe
can take cognizanceof is long enough to give the new mean
of variation,n,a lot of generationsin which to multiplyand
deposit its representativefossils. Of course,this would be
onlythe actionof naturalselectionupon "preformed " varia-
tionsin thosecases whichdid notinvolvepositivechanges,in
structureand function, acquiredin ontogenesis; but in so faras
such ontogeneticadaptationswere actuallythere,the extent
ofdifference ofthe n mean fromthe x mean wouldbe greater,
and hence the resourcesof explanation,both of the sudden
prevalenceof the newtypeand of its discontinuity fromthe
earlier,would be much increased. This additionalresource,
then,is due to the " OrganicSelection" factor.
We seem to be able also to utilize all the evidence usually
citedforthefunctionaloriginofspecificcharactersand group-
ings of characters. So far as the Lamarkians have a strong
case here,it remainsas strongif OrganicSelection be substi-
futedforthe " inheritanceof acquired characters." This is
especiallytruewhereintelligentand imitativeadaptationsare
1896.] WaterCurrent
in Cucumber
Plants. 451

involved,as in the case of instinct. This " may give the

reason,e. g.,that instinctsare so oftencoterminouswith the
limitsofspecies. Similar structuresfindthesimilar uses for
their intelligence,and they also find the same imitative
actionsto be to theiradvantage. So the interactionof these
consciousfactorswith natural selectionbringsit aboutthat
the structuraldefinitionwhich representsspecies, and the
functionaldefinition whichrepresents instinct,largelykeep to
the same lines" (ref.5).
6. Itseems proper,therefore,to call theinfluenceofOrganic
Selection"a new factor;" forit givesa methodof deriving
the determinategains of phylogenyfromthe adaptationsof
ontogenywithoutholdingtoeitherofthetwocurrenttheories.
The ontogeneticadaptationsarereallynew,notperformed;and they
are reallyreproducedin succeeding althoughnotphysi-
generations,
cally inherited.
(To be continued.)

THE PATH OF THE WATER CURRENT IN CUCUM-

BER PLANTS.
BY ERWIN F. SMITH.
(Continued from page 378).

2. UPWARD MOVEMENT OF ONE PER CENT. EOSINE WATER

THROUGH CUT STEMS PLUGGED WITH GELATINE.

In all of theseexperimentsa somewhatstiffgelatinewas

used (15 per cent.) to securea relativelyhigh meltingpoint
(about 270 C.) and thiswas tingedwithIndia ink,so that the
locationof the gelatine plugs insideof the vessels could be
determinedaccuratelyon cross section. Both substances
being as far as has been determinedinertto the plant,it is
notlikelythattheycould have in any way injuredthe carry-
ing capacityof the walls of the vessels.2
2RecentlyDixon and Joly (Annals of Botany, Sept., 1895, p. 403) have raised
some objectionsto this view, butit cannotbe said thattheyhave fullyestablished
theircase.