Vous êtes sur la page 1sur 24

Laticifers: An Historical Perspective

Author(s): Paul G. Mahlberg

Source: Botanical Review, Vol. 59, No. 1 (Jan. - Mar., 1993), pp. 1-23
Published by: Springer on behalf of New York Botanical Garden Press
Stable URL: http://www.jstor.org/stable/4354199
Accessed: 03/03/2009 16:29
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the
scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that
promotes the discovery and use of these resources. For more information about JSTOR, please contact support@jstor.org.

New York Botanical Garden Press and Springer are collaborating with JSTOR to digitize, preserve and extend
access to Botanical Review.


VOL. 59


No. 1


Laticifers:An Historical Perspective

Department of Biology
Indiana University
Bloomington, IN 47405, USA

I. Abstract.1....
II. Introduction..
III. TerminologyPertainingto LaticiferStructure
IV. Conceptson Formation
A. IntercellularSpace
Concept- .
..--B. CellularConcept..5
V. Classificationand Distributionof Laticifers
VI. Originand Development of the NonarticulatedLaticifer
A. Euphorbia-typeLaticifer.11l
B. Variationsof the Euphorbia-typeLaticifer
VII. Mode of Growthof the NonarticulatedLaticifer.-.
VIII. Wall Structureof the NonarticulatedLaticifer
IX. Cytologyof the NonarticulatedLaticifer
X. Summary- .------------------------.--..
XI. Acknowledgments...20
XII. LiteratureCited..





I. Abstract
This reviewdescribesthe developmentof the laticiferconcept,with emphasisupon
the nonarticulatedtype, from early observationsof plant exudates and "juices"to
the presentationof laticifersby Esau(1953). Classicalwritersand herbalistsdescribed
practicalapplicationsof these substances.With the advent of the microscopeearly
investigatorsbelieved that these substancesoccurredin structurespresent in most,
if not all, plants and, wrongly,equatedthese structuresto the circulatorysystem in
animals. Introductionof the term, latex, into botany derived from its early use as a
term for a blood component by physicians,and not for analogy to milk. However,
the originof the terms, laticiferand laticiferous,remainsuncertain.Initial studies of

Copies of this issue [59(1)] may be purchasedfrom the Scientific PublicationsDepartment,The New York BotanicalGarden, Bronx,NY 10458-5126 USA. Please inquireas to prices.
The Botanical ReviewS59: 1-23, Jan.-Mar., 19931
? 1993 The New York Botanical Garden


laticiferswere markedby the controversyof whetherthey representedintercellular

spacesor elongatedcells. Confirmationoftheir cellularcharacterled to the designation
of nonarticulatedand articulatedlaticifers.Nonarticulatedlaticiferswere shown to
ariseduringearlyembryogenyin some plants.The ontogeneticoriginof the articulated laticiferwas unclearto earlyworkers,but new laticifersweredetectedto be formed
by cambiumactivity.Nonarticulatedlaticifersweredescribedto develop by intrusive
growth whereby tips of the cell penetratedbetween adjacent cells. The coenocytic
condition of the nonarticulatedlaticiferresultedfrom nucleardivisions alongthe cell
positioned in the growthregion of the shoot and the subsequentdistributionof the
daughternuclei along the length of the cell.

Die vorliegendeUbersichtbeschreibtdie Entwicklungdes Milchrohrenkonzeptes,
beginnendmit den friihen Beobachtungenan Pflanzenausscheidungenund "Pflanzensaften"bis hin zu ihrer Darstellungbei Esau (1953). Dabei stehen ungegliederte
Milchr6hrenim Vordergrund.Die klassischenSchriftstellersowie die Verfasserder
Krauterbiicherhaben die Nutzanwendungendieser Stoffegeschildert.Mit der Erfindungdes MikroskopswurdenfriiheForscherzu der Annahmeverleitet,daBderartige
Stoffein Strukturenvorkamen,die den meisten,wennnichtallenPflanzengemeinsam
seien. Diese Strukturenwurdendann, falschlicherweise,mit dem Kreislaufder Tiere
homoligisiert.Die Einfuhrungdes Begriffs"Latex"in die botanischeTerminologie
beruhtauf der friiheniirztlichenVerwendungdieses Begriffsfureinen Blutbestandteil,
und nicht auf einer Analogie zu Milch. Die genaue Herkunft der Bezeichnungen
"Milchr6hre"und "Milchsaftfuhrend"bleibt jedoch im Dunkeln. Erste Untersuchungen an Milchr6hrenwaren von der Kontroversegepriigt,ob es sich um sehr
starkgestreckteZellen oder um HohlraumezwischenZellen handelt.Mit der Bestatigungder zellularenNatur der Milchrohrenfand ihre Einteilungin gegliederteund
ungegliederteMilchrohrenstatt. Es konnte gezeigt werden, daB ungegliederteMilchrohrenschon in den friihen EmbryonalstadienmilchsaftfuhrenderPflanzen angelegtwerden.Die ontogenetischeHerkunftgegliederteMilchr6hrenkonntevon den
friihenBearbeiternnicht geklartwerden;sie stelltenjedoch fest, daBneue MilchrohrenzellendurchkambialeAktivitatabgegliedertwerden.Es wurdeauch beschrieben,
daB ungegliederteMilchrohrenwaihrenddes Wachstumsin bereits vorhandeneGewebe eindringen,wobei ihre Zellspitzensich zwischen die ZellwiindezweierbenachbarterZellen drangen.Die coenocytischeNatur ungegliederterMilchr6hrenkommt
durchKemteilungenan der Spitze der Milchrohreund damit des Sprosseszustande,
wobei sich die Tochterkernenach ihrer Abgliederungentlang der gesamten Zelle
II. Introduction
The conspicuousmilky content in certainplants has attractedcuriosityfor many
centuries. Early scholars recognizedthat the colored, milky substances,as well as
those of a mucilaginousor resinouscharacter,were restrictedto particularplants. In
the classicalliteraturetherewereoccasionalreferencesto the collectionof theseplants
for their peculiar contents. Presumablysuch substances were utilized for various
practicalpurposes.Theophrastus(1916) referredto the milky juices of the spurges
which were collected for medicinal purposes.He also referredto the juices of other


plants, especiallyin the Apocynaceae(NeriumoleanderL.), which had become well

knownfor their poisonous character.Theophrastusvery broadlysuggestedthat these
plantjuices were a fundamentaland essential component of all plants, an interpretation readilyacceptedby his students.However, the derived implicationthat such
substanceswere not of a secondaryorigin, either secretoryor excretory,provided
the basis for a prolongedinterpretativecontroversy.
Earlystudies were restrictedto superficialdescriptionsof plant material,such as
exemplified in herbals, while anatomical studies could be pursued only after the
improvement of the compound microscope by Hooke in 1665. One of the initial
topics investigatedby earlymicroscopistswas the natureand the distributionof these
juices. However,the inadequaciesof the firstmicroscopes,supplementedby the vivid
and imaginativedescriptionsof early authors,resultedin the formulationof rather
contradictoryanatomicalconceptsof the form and developmentof structureswhich
contained milky contents or other "plantjuices."
Malpighi(1901), in his classicalwork on plant anatomy,describedplant material
as composed of two structuralunits; the utricleswhich constitutedthe plant body,
and the tubes which were distributedwithin the utricularbody. In the various plant
materialsinvestigated,such as fig, cypress, celery and elderberry,he distinguished
two types of tubes;the trachea,which includedthe presentvasculartissues, and the
vasapropria,which contained the milky, resinous or mucilaginoussubstances.
It appearedthat Malpighiconsideredall plants to possess the vasa propriawhen
he stated:"Thereare several kinds of vessels (tubes)in plants both in the bark and
in the wood in additionto the vasapropria."Further,he contendedthat the contents
of the vasapropriamay not alwaysbe superficiallyvisible. The liquidwhich appeared
upon the surfaceof a fresh wound of some plants constituteda part of the essential
plant sap containedin the vasa propria.
The nature of the vasa propria was somewhat less certain for Grew (1682), a
contemporaryof Malpighi.He likewise distinguishedtwo types of vessels within the
parenchymatousplant body: the air-vessels, actually the vascular conducting elements, and the lymphatic vessels which included the vasa propriaof Malpighiand
certainother structuralcomponents of phloem. The structureswhich he interpreted
as lactiferous,resiniferousand mucilaginousvessels were includedin his categoryof
lymphaticvessels. Grew (1682) describedthe occurrenceand distributionof lactiferous vessels in several plants, including dandelion, endive, Scorzoneraand sumac.
However, he did note that the lactiferoussubstance was not present in all of the
plants which he studied.

III. TerminologyPertainingto Laticifer Structures

Grew's interpretationof the lactiferoussubstancewas derived undoubtedlyfrom
its analogyto milk in animals. It was similarto milk both in color and coagulability.
However,this relationshipwas not acceptedby subsequentworkers.Rather,the term
latex (Latin,meaningfluid or liquid) supersededGrew'sterm, becomingestablished
among English-speakingphysicians as early as 1662 (Chandler,1933). The author
who initially suggestedthe applicationof it or the adjectivalform, laticiferous,with
referenceto plant material,is difficultto determine.Schultz(1839), a Germanphysician, was one investigatorwho employed the word, latex, in his botanical publications. In the field of medicine, it was used in referenceto the characterof blood.
"Her blood appearedof a good texture, otherwisethan giving off a little more than


its due proportion of latex" (Spry, 1767). Like blood, the latex in the plant was
thoughtto be containedin a vessel system. Latexalso coagulatedupon removalfrom
the plant, as did blood. Schultz communicatedhis observationsto Lindley (1848),
who publishedthem in some detail. In this work, Lindley also referredto the milky
substanceas latex. The termlatex remainedwell entrenchedin the botanicalliterature
thereafter.Duringthe first half of the 18th century,many of the anatomicalstudies
upon plants were performedby investigatorswith some medicaltraining.Thus, it is
understandablehow any term with medical implications would be injected into
The term laticiferalso has appearedin the literature(Esau, 1953; Jackson, 1928),
and was more convenient than such terms as laticiferousvessel or laticiferousstructure. Since the composition of latex was quite variable,it was difficultto define as a
substance (Bonner & Galston, 1947). Historically, latex was characterizedby its
capacityto coagulatewhen removed from the plant;its compositionwas completely
unknown. Superficially,latex may appearnearly clear (Nerium),or white and very
turbid (Euphorbia).It also may be colored, either yellow-brownas in Cannabis,
yellow-orangeas in Papaver,or red as in Sanguinaria.However, these observations
providedno data on the chemical composition of latex. In early studies of laticifers
the most preciselyidentifiedsubstancefoundin latexwas starch(Hartig,1843;Potter,
1884; Trecul, 1865a).

IV. Conceptson Formationof Laticifers

The descriptionand interpretationof the structures,that Malpighidesignatedas
the vasa propria, and the lactiferous vessels of Grew, provided the basis for the
development of two distinctly differentconcepts of these structures.These may be
designatedas the intercellularspace concept and the cellularconcept.

Some investigatorssupportedthe theorythatthe colored,resinous,or mucilaginous

substanceswere containedwithin vessels or intercellularspaces (Anonymous, 1846;
Bernhardi,1805;Link, 1824;Meyen, 1838;Mirbel, 1815;von Mohl, 1844;Schleiden,
1849; Schwann, 1839; Sprengel, 1817; Treviranus, 1835). Most often these investigators attempted to relate the supposed structuresthey observed to a particular
function within the plant.
The greatlength and extensive distributionof the laticiferoussystem in the plant
induced several investigatorsto compare it with the circulatorysystem in animals
(Mariotte,1717;Meyen, 1838;Schultz,1839,1841; Trecul,1860;Unger, 1840;Wolff,
1869), which was firstdescribedin 1628 by Harvey (1941). Schultzpresenteda very
imaginativeinterpretationof the latex system. The dense, colored latex of the plant
body correspondedto blood of animals, whereas the plant sap was equivalent to
lymph in animals. Althoughlatex was not presentin all plants, Schultzthoughtthat
he had observed it in the majorityof plants he investigated.He interpretedit, like
blood, to consist of a coagulum, that coagulatedupon exposure to air, and also a
liquid serum. Latex was derived from the sap of wood and, afterrisingin the wood,
the sap was introducedinto the leaves whereit was subjectedto a processof"elaboration," whereuponit was deposited in the laticiferousvessels as latex. Subsequently,
it moved downwardin the vessels which were distributedwithin the bark. During


its movement the latex supposedlypermeatedall the living tissues, providingthem

with nutritionalsubstances.Upon exhaustionof these materials,the latex returned
to the wood, as sap, whereuponthe circulatorycycle was repeated.
Movement of latex and sap was attributedto both externalfactors(heat and light)
as well as internalfactors (contractionand dilation). Like Malpighi(1901), Schultz
(1841) thought that he could observe the cyclosis of latex which resultedfrom the
contraction and dilation of the walls of the laticiferous vessels. Supposedly this
peristalticmovement was equivalent to the heartbeatsin animals.
Severalinvestigatorsinterpretedthe laticiferousstructuresas intercellularsecretory
cavities (Anonymous, 1846; Link, 1837; Mirbel, 1815). Ratherthan representinga
circulatorysystem,they regardedlaticifersas "reservoirsin whichthey collectedtheir
own juice" (Link).Mirbelexpandedupon this hypothesisfrom his own observation,
describingthe "cannauxsecretoires"as possessing a very fine limiting membrane.
He was able to identify this membrane consistently in various members of the
Apocynaceae,Asclepiadaceaeand Euphorbiaceaethat he investigated.However,the
membrane appearedto be evident only in mature portions of the plant and was
interpretedto arise after the formationof the secretorycanals.
Presenceof a membranewas substantiatedalso by an anonymous writer (1846)
who surveyedgenerafromfamiliesbelieved to containa laticiferoussystemincluding
the Apocynaceae,Araceae,Asclepiadaceae,Campanulaceae,Caprifoliaceae,Cichoriaceae,Cucurbitaceae,Euphorbiaceae,Lobeliaceae,Moraceae,Papaveraceae,and Urticaceae.The membrane,accordingto this author,lined the entireintercellularcavity.
Althoughnot evident duringthe initial developmentof these cavities, it was present
in later stages. The membrane,which became increasinglythicker in older canals,
was interpretedto be deposited along the inner surfaceof the canal by the adjacent
Some investigators,although adherents of the general cellular theory of plants,
were unwillingto ascribe a cellularnatureto the laticiferousstructures(von Mohl,
1844, 1852; Schleiden, 1844, 1849; Schwann, 1839). A quotationfrom von Mohl is
indicative of the uncertaintywith which he viewed them. "In the majorityof plants
containingmilkyjuices, these canalsare lined with a specialmembraneand are then
calledmilk vessels, but can scarcelybe separatedfrom merecanalsdestituteof proper
membranesrunningbetweenthe cells, since truelatex is formedin the latterin many
plants, as in Rhus."
A similar uncertaintywas evident in Schleiden'sinvestigation.However, he did
consider the resemblancesthat these structuresshared with cells: "The vessels of
latex sap which arisewith their own membranecannotbe tracedbackby observation
to cells. Their origin is obscure,in the developed state they are similarto elongated
branchedcells." The incongruitiesin descriptionsof the laticifersstimulatedadditional investigationsand the formulationof new ideas.

An elemental cellular concept of the laticifer was introducedquite early in the

investigationof the plantbody. Wolff(I 869) formulateda theoryintendedto explain
the formationof utricles(cells). Essentially,he maintainedthat the youngestpart of
the plant, the punctumvegetationis,consisted of a gelatinousmatrix. The latterwas
saturatedwith a nutrientsap-like substance.Small drops of this sap very gradually
increased in size, resulting in the formation of the utricle or cell. The gelatinous


matrix representedthe cell wall. Elongatedvasa propriawere producedby the longitudinalextension of particulardrops of the nutrientsap.
Plant growthwas the result of continuedformationof new utriclesor cells among
those alreadyformed. The cell wall matrix was representedas a homogeneoussubstance, precludingthe existence of any intercellularspaces. This conclusion was
somewhatin contrastto that of Grewwho recognizedintercellularspaces.However,
Grew was not certainof their relationshipto his "lactiferousvessel."
The cellular nature of Malpighi's vasa propriawas expounded by several early
investigatorsand was contemporarywith the intercellularspace concept (Moldenhauer, 1812; Trecul, 1865b; Unger, 1847; Wolff, 1869). However, the concepts expressedby these investigatorswere quite dissimilar.
A more accurateinterpretationof cellulararrangementwithin the plant body was
presentedby Moldenhauer(1812) who employed a macerationtechniqueto isolate
individualplant cells. He theorizedthat if cells could be isolated, then each cell must
possess its own wall. Similarly,any two adjacentcell cavities must be separatedby
two walls. Utilizing this theory he investigatedand redescribedthe vasa propriaof
Malpighiwith some degreeof accuracy,referringto them as cells. He describedthe
vessel-likenatureof the laticifersin Musa, Asclepiasand Chelidonium.However,he
did not considerthe resin canals of Pinus to be similarto the laticifer.Moldenhauer
emphasizedthat the presenceof a discretelayer of cells surroundingthe resin canal
and the presenceof a special membranelining the inner side of the canal suggested
that the canals were not related to laticifers.
The significanceof Moldenhauer'stheorywas not recognizedimmediately,but the
theory did aid in defining the cellular nature of plant tissues in general. Several
investigatorssuggestedthat laticiferousstructureswere very elongatedcells (Dippel,
1865;Faivre, 1868;Hanstein, 1864;Hartig,1862;Schacht,1851, 1856;Unger, 1840,
1847; Vogl, 1863). The formationof the laticiferousvessel was vividly describedby
Ungerfrom observationsthat he made on Ficus benghalensisL. Vessels were formed
from superimposedrows of cells. The greatlengthof these vessels was attainedupon
resorptionof transversewalls that separatedthe cellularcomponents.He noted that
the walls of these vessels were initially quite delicate. During their subsequentdevelopment,the cells laid down additional,but ratherirregular,thickeningsupon their
Laticiferswere dispersedthroughoutthe plant body and, accordingto Unger, the
vessels were joined into a complex system by means of lateralbranches.Although
correct in many essentials his description of the formation of the laticifer in this
plant was proven later to be incorrect.Nevertheless, since his interpretationwas
published during the period in which the intercellularspace concept was widely
recognized,it did stimulate furtherinvestigationson this topic.
Schacht(1851) investigatedthe laticiferousstructuresin various generaincluded
in the Apocynaceae( Vinca),Asclepiadaceae(Hoya,Gomphocarpus),
Caricaceae(Carica), Cichoriaceae(Lactuca, Sonchus), Euphorbiaceae(Euphorbia),Papaveraceae
(Papaver,Chelidonium).In all instances he found that laticifers were of a cellular
origin. He describedthem as resultingfrom the fusion of many or a few cells into a
vessel. Since he was unable to find the laticiferousvessels formed into an anastomosing system in all of the plants that he investigated,he dismissed the possibility
that they served any circulatoryfunction similar to that presentin animals.
Schacht(1851) clearlydescribedthe formationof the laticiferin Caricaas arising
from rows of superimposedcells, the transversewalls of which were resorbed.The


formof the laticiferpresentin EuphorbiaandHoya was moreobscure.Its resemblance

to bast cells presentin surroundingtissues induced Schachtto considerit simply as
a branchedbast cell (also Pitra, 1860). He had become even more assertive on the
natureof the laticifer,in Euphorbia,when he stated:"Throughmany investigations
I have shown the non-existenceof truelaticiferousvessels. The latterform no anastomosing system. They are nothing more than branchedbast cells bearinglatex, with
their ends completely closed." Schacht's correlation of the laticifer with another
cellularentity, such as the bast cell, was not unique. It also was construedto correspondto sieve tubes (Dippel, 1865; Hartig, 1862; Vogl, 1863).
Hanstein (1864) supportedmany of the observationsthat Schacht had made on
his materials. He expressed no doubt of the cellular nature of the laticifer. In his
survey, Hanstein observed that in several families (Alismataceae,Araceae, Campanulaceae, Caricaceae,Cichoriaceae,Lobeliaceae, Papaveraceae)the laticiferous
vessel was formed by the fusion of adjacentsuperimposedcells.
Hanstein did not considerthe type of laticiferpresentin the Apocynaceae,Asclepiadaceae,Euphorbiaceae,or the Urticaceaeto be a bast cell or a phloem element.
Such factors as the very early differentiation,great length, irregulardistribution,
branchinghabit, and the moderatelythickened cell wall suggestedto him that the
laticiferwas not a latex-bearingbast cell. He did admit that transitionstagesbetween
the thick-walledbast cell and the laticifermay occur. He consideredthe laticiferto
be quite distinct. However, he thought that laticifers,collectively, formed a closed
Although connections may exist between the individual laticifers,no communications were evident with other cells. Nevertheless, Hanstein did not consider the
laticiferoussystem as a distinct tissue; he was unable to ignore the superficialresemblance which it did have with the components of the bast system. Thus, he
concludedthat the laticifersrepresentedconstituentsof the bast system.
With the publication of additional investigationsupon the laticifer,all of which
supportedthe cellular concept (Dippel, 1865; Faivre, 1868; Hartig, 1862; Unger,
1847; Vogl, 1863), the intercellularspace concept of the laticifer was finally supersededby the cellulartheory.

V. Classificationand Distributionof Laticifers

Recognition of structuraldifferencesamong laticifers in laticifer-bearingplants
contributedto the development of several classificationschemes. Several authors
attemptedto classify laticifersby their form (David, 1872; Hanstein, 1864; Hartig,
1862; Mayus, 1905; Trecul, 1865c, 1866; Unger, 1846, 1858; Vogl, 1866). In one
classificationlaticifersweredesignatedas the Y-form to distinguishbetweena simple
branchingpatternand the H-form which representedsupposedfusion of two vertical
branches.More complex patternsinvolving unicellularand multicellularcomplexes
of cells and tubes were proposedby Gaucher(1902).
Hartig (1862) presented one of the initial attempts to interpretand classify the
anatomical variability that he had observed in several laticiferous plants. While
endeavoringto investigatethe movement of latex in variousplants,he describedthe
latex system in Acerand Chelidoniumas composed of articulatedtubes (gegliederten
Rohren)that he contrastedwith those structuresexemplifiedin the Euphorbiaceae,
and referredto as nonarticulatedvessels (nichtgegliedertenMilchgefasse).
Utilizing maceratedmaterials,he describedthe articulatedlatex tube as composed


of a row of superimposed cells in which the cross walls of component elements

became perforatedduringthe processof differentiation.This interpretationwas similar to that presentedby Unger at an earlierdate (1847). The nonarticulatedlatex
vessel, in contrast, was a very much elongated cell with no detectable cross walls
along its entire length.
Hanstein concluded from his extensive survey of laticiferousplants that characteristicdifferencescould be detectedbetweenthe laticiferspresentin variousfamilies.
In the Cichoriaceae,Campanulaceae,Lobeliaceaeand Caricaceaehe observed that
anastomoseswere presentbetween adjacentarticulatedvessels. The laticiferoussystem appearedvery much like a closed networkof cells distributedin the extracambial
regionof the shoot. Within the Papaveraceae(Chelidonium,Papaver,Sanguinaria),
Hansteinfound that the anastomoseswerequite infrequentand restrictedto laticifers
in leaves, cotyledons and carpelsof the ovary.
Chauveaud(1891) presenteda detailedclassificationof the formsof laticifersbased
upon more specific anatomicaldifferencesrecordedin various generaregardlessof
theirtaxonomicposition. In his interpretationthe laticiferoustissueswere composed
either of cells or tubes. He subdivided the tubes into two types: 1) a continuous,
nonarticulatedtube that arose either originally in the embryo (original form), or
during the post-embryonicdevelopment (subsequentform); and, 2) an articulated
tube consistingof eitherseparate,fused,or anastomosingelements.The cellularform
also was subdivided accordingto its disposition, and was classifiedinto types that
were arrangedeither in series or occurredas isolated cells (Table I).
The distinction between the forms of laticifers was not as sharp as might be
suggested in the tabular summary of Chauveaud'sclassification.The number of
generathat had been investigatedin detail was still minimal when comparedwith
the approximately800 or more generaincludedin the families possessingnonarticulated laticifers.
It was De Bary(1884), however,who adoptedHartig'sterminologyand established
the two categoriesof laticiferoustubes: the articulatedtype and the nonarticulated
type. The convenienceand applicabilityof these terms was widely acceptedby subsequentauthors(Esau,1953;Foster,1949;Haberlandt,1914;Sperlich,1939;Tschirch,
Esau (1953) elaboratedupon the classificationof the laticifer,utilizing the variations observedamongthem. Nonarticulatedlaticifersweresubdividedinto two forms:
those in which the cells developed as individual elongatedtubes were termed nonarticulatedunbranchedlaticifers;and, those in which the cells branchedrepeatedly
duringtheir development were termed nonarticulatedbranchedlaticifers.
Articulatedlaticifers also were subdivided by Esau. If no anastomoses occurred
between adjacenttubes in the plant body they were designatedas nonanastomosing,
in contrastto the anastomosingform in which lateralanastomosesdid occur (Table
Severalstudies suggestedthe existenceof variationsfrom these forms.These modifications may occur either in the same family, in the same genus, or even in the
same individual. One such modification was exemplified by the capacity of the
component cells of certain articulatedlaticifers(Lactuca, Chelidonium)to undergo
a limited amount of intrusive growth. Protuberancesthat developed on these cells
intrusively forced their way between the adjacentcells until they came in contact
with anotherlaticifer.Resorptionof the cross walls at the point of contact resulted
in a direct communicationbetween the two laticifertubes (Calvert, 1887; De Bary,

Table I
Classification of Laticifers (Chauveaud, 189 1)


Beginningin embryo, traversi

the plant:Euphorbia,Croton




Occurringas one long vessel, o

one from anotherby transve


Occurringas one long vessel, o

transversewalls is more or le


Occurringas one long vessel, o

cross walls is complete. In ad
are presentbetweenthe adja











Table II
Classificationof Laticifers(Esau, 1953)
Apocynaceae,Eucommiaceae,Moraceae,and Urticaceae(in part).
Apocynaceae,Asclepiadaceae,Euphorbiaceae(in part),and Moraceae.
Convolvulaceae,Liliaceae,Papaveraceae,Sapotaceae,and Urticaceae (in part).
Campanulaceae,Caricaceae,Compositae(Cichoriaceae),Euphorbiaceae (in part),and Papaveraceae.

1884; Fraser, 1931; Parkin, 1900; Scott, 1884). Thus, intrusive growth may not be
restrictedto the nonarticulatedtype of laticifer.
It should be noted also that both the articulatedlaticifer, as in Hevea, and the
nonarticulatedtype, as in Euphorbia,occurredin the Euphorbiaceae(Schaffstein,
1932). It had been reportedby Schaffstein(1932) that both laticifertypes could occur
in the same plant as was the case in Stapelia and Trichocaulon(Asclepiadaceae).

VI. Origin and Developmentof the

Many of the investigationsupon the two types of laticiferswere conductedafter
the initial formation of the structureshad occurred within the plant body. The
presenceof both types was readily apparentin the axes and meristems of both the
seedlingand matureplant. How these structuresinitiallyarosewithin the tissueswas
not understood.Trecul (1865c) and Faivre (1866) brieflydescribedthe presenceof
laticiferswithin mature embryos of the Asclepiadaceae,Euphorbiaceae,and Compositae, but neitherinvestigatedthe developmentalaspects in detail. Dippel (1865)
and David (1872) unsuccessfullyattempted to explain this fundamentalaspect of
their development,but neitherworkerbecame awareof the significanceof the laticifers alreadypresentin the embryo of the mature seed.
Dippel contendedthat all laticifersarosefromthe coalescenceof cells. He observed
that the tubes could be readilyfollowed into the young meristemswherethey ended
abruptly.Near the ends of the so-called nonarticulatedlaticifer,he thought he saw
the remains of wall septa within these tubes (Ficus carica L., Euphorbiasplendens
Boj.).It appearedto Dippel thatthe processof cell fusiontook placein the meristematic zones and occurredvery rapidly.This was in contrastto the septathat werereadily
detectablealong a considerablelength of the laticifersin the Cichoriaceae,Papaveraceaeand Campanulaceae.He provided no explanationfor this structuralincongruity, nor indicated whether all laticiferswere derived by a similar process of cell
Dippel (1865) ascribed the origin of new laticifers to parenchymatouscells on
either side of the vascularbundles in the shoot. Althoughhe observed branchesof
the laticifersfrom the stem extendinginto the petiolarbase of leaves, he contended
that the laticiferoussystem within the laminae was independentof that within the
stem. Rather, the laticiferoussystem within the blade was derived from parenchymatous cells which differentiatedinto laticifercells duringthe development of each



lamina. These new laticifer initials, by means of coalescence with adjacent cells,
progressivelydeveloped into a ramifiednetworkof tubes that extended throughout
the entire leaf.
Upon reinvestigatingsimilar generautilized by Dippel, David (1872) was not in
complete agreementwith Dippel's conclusions.Accordingto David the laticifersin
the Apocynaceae, Asclepiadaceae,Euphorbiaceaeand Moraceaewere single cells
(nonarticulated),which "elongatedto a significantlength by means of active and
passive stretchingand also by means of branchinginto the intercellularspaces."The
laticiferouscells couldbe presentin both the cortexand the pith (Euphorbiasplendens,
Ficus elastica L., Nerium oleander,Hoya carnosa R. Br.) or were confined to the
cortex (E. cyparissiasL., F. carica).
David (1872) maintainedthatnew laticiferswereformedprogressivelyfromcertain
cells of the groundtissue below the terminalmeristemduringthe growthof the plant.
Each new laticiferfrequentlybranchedas it elongated,but no anastomosescould be
observed between adjacentbranches.He was not certain of their relationshipwith
the vascularbundles. In the stem the laticiferswere randomlydistributed,while in
the petiole and the blade of the leaf they were associatedwith the vascularstrands.
David, in contrastto Dippel, observed that the brancheswhich extended into the
petiole fromthe stem continuedinto the blade of the leaf to forma continuoussystem
(Euphorbia).However, in Ficus, Nerium, and Hoya it did appear to him that the
"leaf specific"laticifer,as describedby Dippel, did occur.Mayus (1905), in contrast
to Dippel, recognizedthat the entirelaticiferoussystem of the leaf was a continuation
of branchesfrom the stem. Tips of laticifer cells were observed in these plants to
penetrateamong mesophyll cells to and between radial walls of epidermalcells to
the cuticle of the leaf (Chauveaud,1891; Gaucher, 1902; Groom, 1889).
Both Dippel (1865) and David (1872) observed that the laticifercould be distinguished from adjacentcells in the embryo and meristems well before the cellular
elements of either the xylem or phloem became identifiable.David found no indication in his material that new laticifers were produced by cambial activity, as
suggestedby Dippel.

Emphasishere is placed on the nonarticulatedlaticiferin Euphorbiabecausethis

genus had been studied more intensively by several investigatorsthan any other
genus and, thus, can be employed as a model for understandingthis laticifersystem
in other euphorbiaceousgeneraand other laticifer-bearingfamilies.
The divergentconclusionsderivedfrom investigationson the originof the laticifer
in the shoot stimulated a reconsiderationof Trecul'searlierstatement that he had
observed laticifers in mature seeds of Asclepias cornutiDecne. and Euphorbialagascae Spreng.These observationswere readilyconfirmedby Schmalhausen(1877)
and Chauveaud(1891). Both investigatorsintensively studied the origin and development of the laticiferin embryosof Euphorbia.They found that laticiferswere first
distinguishableshortly after initiation of the cotyledons. Only a relatively small
numberof laticiferinitials, such as four,eight or twelve, wereformedin each embryo.
The only characteristicswhich distinguishedthese initials from adjacentcells were
their largersize and ratherrefractivewalls. The latterappearedto become thickened
or swollen in appearance.
Chauveauddefined these cells in the following manner:"In order that these be



distinguishedby theirorigin,let us call these the initial cells of the laticiferoussystem,

or more brieflythe initial cells, or even more simply the initials."
These initials occupied a position immediately below the primordiaof the cotyledons. Chauveaudtermed this region the cotyledonarynode. The number of cells
which develop into initials appearedto be constantwithin a particularspecies. However, the numberof initials varied between species and genera.Chauveaudreported
in E. engelmannii Bois. that only four initials were situated individually at four
symmetricalpoints which coincided with the position of the vasculartraces in the
cotyledonaryplane. In other species (E. exigua L., E. peplus L.) there were eight
pairs at these same points, whereas in E. segetalis L. a largernumber of initials
formed symmetricalarcs at the position of the vasculartraces.
In other species,as E. myrsinitesL., he describedtwo arcsof initials which formed
a semicircle on either side of the axis, and were interruptedat the two extremities
of the trace to the cotyledonaryplane.
In some species the entire layer of cells located in the equatorialplane was transformedinto initial cells, the latterthen forminga complete circleat the cotyledonary
node (E. portandicaL., E. lathyrisL., E. falcata L.).
Chauveaud described the initials as arising from the pericyclic tissue, whereas
Schmalhausen(1877) found it difficultto associatetheir originwith a specifictissue,
as evident in the followingtranslationfrom his account:". . . the line which sharply
delimits the pleromefrom the periblemat its upper(cotyledonary)end meets directly
at these cells, and these cells often appearto be wedgedin betweenthe pleromeand
periblemcells which borderthem below."The difficultyof associatingnonarticulated
laticiferswith a particulartissue also was evident to Blaser (1945), who concluded
that they were outside the vascularsystem.
BothChauveaudand Schmalhausenobservedthat, as the youngembryoscontinued
to grow, the laticiferinitials increasedin length. The upper and lower ends of each
initial appearedto undergoapical growth,forciblypushingtheir tips between other
cells. In this mannerthe upperprotrusionwhich Chauveaudtermedthe cotyledonary
tube grewinto the developingcotyledon. The protrusionat the lower end of the cell
extended downwardlytowardthe tip of the radicle, formingthe centraltube.
Chauveaudcontendedthat each initial also producedlateralbranchesat the level
of the nodal plane. Growingmore or less horizontallyalong the intercellularspaces,
these branchesdevelopedfrom the initialsduringthe formationof this nodal network
and grewinwardand upwarduntil the tips nearlyreachedthe shoot meristem.These
brancheshe termedthe plumuletubes. Likewisebranchesdevelopedoutwardlyfrom
the plexus and grew into the cortical zone. These tubes, termed the cortical tubes,
subsequentlyturned downwardand grew toward the tip of the radicle along intercellularspaces.Otherinvestigatorsconfirmedthe occurrenceof a branchedlaticiferous system in the embryo (Cameron, 1936; Vreede, 1949).
Thus a clearedmatureembryo of Euphorbiawould appearto be permeatedwith
a laticiferoussystem. At the cotyledonarynode, it would be apparentas a ring of
interwoven tubes from which branches would extend upwardinto the cotyledons
and the shoot apex. Branchesof the laticiferalso would extenddownwardto the root
meristem both along the immaturevascularcylinderand along the outer periphery
of the cortex.
As noted by both Chauveaudand Schmalhausen,the diameterof the laticiferwhen
viewed in transectioncould vary considerably.They noted that the diameter of a
laticiferin the nodal plexus of the embryo may be two or three times the diameter



of the adjacentcells. The diameterof branchesdevelopedfromthe laticiferousinitials

graduallydecreasedthroughouttheir course. At the growingtip their diameterwas
considerablyless than that of the adjacentcells. Walls of these intrusivelygrowing
branchesalso retainedthe capacityto stretch,becausein maturetissues the laticifers
very often were of greaterdiameterthan adjacentcells.
Chauveaud(1891) and Schmalhausen(1877) did not agree with respect to the
occurrenceof fusions between the laticiferinitials duringtheir development in the
embryo of Euphorbia.Schmalhausencontendedthat he had observed the fusion of
the brancheswhich developed from the initials. He thought he could detect some
places where a dissolution of the common walls between two contiguousbranches
occurred,resultingin fusion of their protoplasts.Similarly,he describedconnections
between the cortical and central tubes toward the tip of the radicle. Chauveaud,
however, sharplydisagreedwith Schmalhausen.He was unable to detect the fusion
of any adjacent laticiferouscells during their development in the embryo of Euphorbia.Likewise,he did not observeany indicationof fusion in variousothergenera
believed to contain the nonarticulatedlaticifer (Aleurites,Asclepias, Croton,Hura,
Jatropha, Vincetoxicum,and others).
Chauveaudand Schmalhausenmaintained that the entire laticiferoussystem of
the mature plant of Euphorbiaarose from the various branchesproduced by the
initial cells formed within the embryo. No new initials were formed in the shoot
duringgrowth,as reportedby Dippel (1865). Upon activationof the meristemsduring
germination,the variousbranchesformedby the laticiferousinitialsalso commenced
to grow. Chauveaud(1891) contended that the centraland corticaltubes kept their
position in the meristem region of the root by means of intrusive growth at their
tips. He also stated that the cotyledonarytubes, as well as their ramifications,elongated at a rate equal to that of the elongation and development of the cotyledons.
Tips of the plumuletubes also kept pace with the growthof the epicotyl and retained
a position in the meristematiczone of the shoot.
In the internodalregion of the stem the laticiferousbranchesmaintaineda rather
straightcourse,exhibitingvery little branching.Both Chauveaudand Schmalhausen
observed that the tips of each laticifer branchedat each node in the shoot. Some
branchescontributedto the formation of a nodal plexus similar to that formed at
the nodal plane in the embryo. Severalbranchesextendedinto young leaf primordia
on the meristem. These branchesformed the entire laticiferoussystem of the leaf.
Tips of the remaininglaticiferbranchesdeveloped from the nodal plexus were observed to maintain a position in the meristematiczone of the shoot.
Nonarticulatedlaticifer branches in the shoot were not confined to a particular
tissue, but ramifiedthroughoutthe shoot (Blaser, 1945). In the shoot they formed
H- and Y-configurationsreflectiveof the branchingpatternfor the growingcell tip.
Some tips were observed to penetrate to the epidermal layer while others were
detected in contact with the cambial zone.
The presence of laticifers in roots of Euphorbiahad been confirmedby several
other workers (Chauveaud, 1891; Schaffstein, 1932; Schullerus, 1882). Schullerus
emphasizedthat branchesof laticifersdid not enterinto lateralroots until the lateral
root had attained a diameter of 1-2 mm. In roots the laticiferswere usually found
to be smallerin size and more difficultto recognizethan those in the shoot. Those
in the root appearedto containa considerablylowerprotoplasmiccontentalongtheir
length than the laticifertubes in the shoot.
Both Chauveaud and Schmalhausenconfirmed the earlier conclusion of David



(1872) that nonarticulatedlaticiferswerenot formedby cambiumactivity. The entire

laticiferous system arose from embryonal initials and their intrusively growing

The Euphorbiaceaepossess considerablevariation in the form of laticifers.The

genera,Ricinus and Mercurialis(Schaffstein,1932) were reportedto contain no laticifers;Hevea and Manihotcontainedthe articulatedlaticifertype;the generaJatropha and Aleuritesare described (Chauveaud, 1891; Gaucher, 1902; Pax, 1884) as
having both types of laticifers.However, in subsequentinvestigationsonly the nonarticulatedtype was identifiedin the lattertwo genera(Schaffstein,1932; Solereder,
1908). In CeropegiathwaitesiiHook. the nonarticulatedlaticiferwas restrictedto the
pith;branchesdeveloped into primordiaof leaves from a nodal plexus similarto the
one formed in the genusEuphorbia(Schaffstein,1932). The section Phyllanthoideae
was reportedto contain no laticifers(Pax, 1884). Gaucher(1902) contendedthat the
articulatedlaticiferwas the predominantform in the Euphorbiaceae.
StapeliabellaBerger(Asclepiadaceae)was reportedto containboth articulatedand
nonarticulatedlaticifers(Schaffstein,1932). The articulatedsystem developed in the
immediate proximity of vascularbundles, whereasthe nonarticulatedtype was reported to be distributedthroughoutthe groundparenchyma.
In several members of the Apocynaceae( Vinca,Amsonia, Tabernaemontana)no
laticiferswereevident in the embryo(Chauveaud,1891;Schaffstein,1932;Solereder,
1908). Observationsindicated that the nonarticulatedlaticifers arose during postembryonalstages of growth. Presumablythe laticiferousinitials arose from certain
cells in the meristemat the base of a youngleaf primordium.These cells subsequently
elongatedduringgrowthof the shoot.
Laticifersin Cannabisand Humulus originatedfrom cells at the base of the leaf
primordiain the shoot meristem,accordingto Zander(1928), who groupedthem in
the Moraceae.In this respect, laticifer origin was quite similar to that believed to
occur in Vinca. Zander also thought that additional initials developed from other
cells within the primordia themselves, and that these cells subsequentlyramified
throughoutthe lamina. These initials could be distinguishedfrom the adjacentcells
only by their somewhat more dense protoplasmand large nucleoli. Growth of the
initial cells appearedto be towardthe meristemonly, perhapsby means of intrusive
growth. Finally, accordingto Zander,the tip of a laticifer penetratedinto leaf primordia and developed along veins of the leaf blade.
In severalgeneraof the Moraceae(Ficus,Dorstenia,Morus, Treculiaand Maclura)
the branchedlaticiferswere reportedto be somewhatcomparableto the Euphorbiatype (Chauveaud, 1891; Schaffstein, 1932; Schmalhausen, 1877). In Maclura,
Schmalhausenobserved the presence of tyloses in the laticifers.Vreede (1949) affirmedthe nonarticulatedstructureof the laticiferin Ficus and describedthe presence
of laticifersin the vascularcambium.
Severalgeneraof the Urticaceaehad been investigatedfor the organizationof the
laticifer system. Laticiferswere reportedto be present in the shoot of the mature
plant of Urtica,but no initials could be detected in the embryo (Chauveaud,1891;
Gravis, 1884; Solereder,1908; Treub, 1880). However, Schaffstein(1932) was able
to detect very delicatelaticifercells in the tissues of the shoot. Presumablythese cells
remainedunbranchedduringtheirgrowthand, as in Vincaand Cannabis,new initials



were differentiatedfrom certaincells of the meristemduringthe growthof the shoot.

Schaffsteinreportedthat laticiferswere not universallydistributedin the Urticaceae
because he was unable to detect them within representativesof the genera,Helxine
and Pilea. However, Guerin (1923) recorded nonarticulatedlaticifers in the pith,
cortex and secondary phloem in both the stem and root of Laportea, Urera and
VII. Mode of Growth of the Nonarticulated Laticifer
The nonarticulatedlaticiferappearedto increasein length by intrusivegrowthof
the cell tip (Chauveaud, 1891; Schaffstein,1932; Schmalhausen,1877; Schullerus,
1882), a processascribedto the fibersome time earlier.The extensive development
of laticifersthroughoutthe plant body provided substantialsupportfor this contention. In addition,when the course of any one branchof an initial was followed along
the shoot towardthe shoot meristem,the axis of the laticiferwas observedto decrease
graduallyin diameteruntil it terminatedas a narrowand rathersharplypointed tip
wedgedbetween two adjacentcells.
The walls of the laticiferconformedcloselyto the outlinesof adjacentcells, pressing
into any of the spaces which occur betweenthe neighboringcells. Consequently,the
wall of a laticiferwas very irregularand jaggedin appearance.Schmalhausen(1877)
comparedthis peculiarintrusivegrowthhabit to that of parasitichyphaeof a fungus
spreadinginto tissues of its host, but differentin that the laticifergrewand branched
only in meristematictissues of the plant.
Schaffstein(1932), in studyingthe growth relationshipsof the nonarticulatedlaticifer in the shoot, maintainedthat there was a very intimate relationshipbetween
the growthof this cell and meristematicactivity of adjacenttissues. The laticiferwas
describedas being influencedby two factors:active growthat the tip of the laticifer
itself, and the influence that adjacent tissues have upon it. Informationwhich he
believed supported his hypothesis of apical intrusive growth in the laticifer was
derived from graftingexperimentson Euphorbiacaput-medusaeL. He attemptedto
determinethe response of the laticiferspresentin the living tissues adjacentto the
graftedsurfaces.If a union of the various living tissues could occur, he speculated
that a similar union between the laticifersof the stock and scion also may occur.
Schaffsteinobservedthat laticifercells in the immediatevicinity of the cut surface
of both stock and scion frequentlydegenerated.However,some of the laticifersmore
distantfrom the cut surfaceremainedprotoplasmic.It was fromthese active laticifers
that Schaffsteinoccasionallyobserved the occurrenceof brancheswhich penetrated
into the callus. Only a few of these branches penetrated through the callus and
approachedthe graft surface. None were observed to enter the primary callus to
make contact with the graftedsurfaces,or to penetratethroughthe graft-unioninto
the adjacentsegment of the shoot. In some instancesthe laticifersformed branches
as they grewinto the secondarycallus. Schaffsteinconcludedfrom these experiments
that mature portions of the laticifers, although surroundedby mature tissues, had
not lost the potential to resume active growth under certain conditions. A similar
conclusion also had been suggestedby Schullerus(1882). The development of new
branchesfrom the laticifersuggestedto Schaffsteinthat this growthwas a response
inherentin the laticiferitself. Branchformationwas indicative of the active growth
at the tip of the laticiferouscell. Schaffsteinalso contendedthat the occurrenceand
the restrictionof growth only to the meristematiccallus tissues supportedhis hy-



pothesis that growth of the laticifer only could occur in meristematictissues, or in

tissue that had regainedmeristematicactivity. The meristematictissues surrounding
the laticiferappearedto exert control over the development of this cell type.
Similarly, Schaffstein(1932) attributedthe growth of the laticifer in the normal
plantto two processes:the growthand the bifurcationof the laticiferin the meristem
suggestedthe occurrenceof active growthat the tip of the cell; and, the increasein
length of the laticifer concurrentwith the elongation of adjacent immature cells
suggesteda secondarystretchingof the walls of all cells. If the tip of a laticiferwas
to retainits position in the meristem,the growthrate of the laticifermust equal the
rate of wall elongationfor all cells below the meristem.However, as he pointed out,
the rate of elongation of the laticifer must be greaterthan the rate of elongationof
any one adjacentcell. This fact is derived from the considerationthat within the
zone of elongationeach verticaltier of cells is increasingin length. Since the laticifer
extends throughall of these tiers, it must elongate at a rate which can be expressed
as the sum of the elongation-ratesin the contiguouscell-tiers.He also employed this
concept to supporthis view that the adjacentcells somehow may influencethe rate
of elongationof the laticifer.Schaffsteinmaintainedthat the meristematiccondition
of the surroundingcells controlledthe rate of growth of the laticifer.Occasionally,
he observed laticiferbrancheswhich terminatedafter growingfor only a short distance. He explainedthis phenomenon in the following quotation:
It is verypossible that these laticiferousducts have had all, or a part, of their
growtharrestedbecausethey remainedbehindthe divisionzone of the shoot, the
only zone within which they were capable of growing.Most of the branchesof
the laticifer which are present in the meristemgrow at a ratheruniformrate.
None have ever been observedto have penetratedinto the distal region of the
meristem immediately below the protodermof the shoot tip. In the dormant
plant, the tips of the laticifersdo not grow.
They grew along the longitudinallyarrangedintercellularspaces leading towardthe
meristem;only occasionalbranchesdeveloped along the intercellularspacesat right
anglesto the plant axis. The directionalgrowthof the branchesaftertheir formation
in the meristem responded in a similar manner. Although branchingoccurredin
variouspartsof the plant, it appearedto be closely relatedto the formationof lateral
organs.For example, branchformationregularlyoccurredat the base of a leaf primordium.One branchsubsequentlydeveloped into the primordiumwhile the other
maintainedits directionof growthupwardinto the region of the shoot apex. Schaffstein attributedthe developmentof laticiferbranchesto stimulifrom the surrounding
tissues. He contended that the leaf primordiummust exert an influence upon the
branch of the nearby laticifer inducing it to grow into that lateral organ. Likewise,
the shoot meristem was the source of a factor, or factors, that induced the second
branchto maintain its course in the meristem.
The phenomenonof branchingin the laticiferwas not clearlyunderstood.Meeuse
(1942) suggesteda mechanismwherebysuch bifurcationsmight be formed. Branch
formationwas precededby division of a contiguousmeristematiccell into two daughter cells. The new wall resultingfrom this division was perpendicularto the longitudinal axis of the laticifer.When a slight protrusionfrom a laticiferwas interjected
between the two primarywalls of the two daughtercells it resultedin formationof
a bifurcation.Subsequentdivisions in the daughtercells facilitated the continued



development of the branch resulting in what Meeuse termed symplastic growth.

Intrusivegrowth,accordingto him, did not occur, in contrastto the interpretation
suggested by other authors (Chauveaud, 1891; Schaffstein, 1932; Schmalhausen,
1877; Sperlich, 1939). As Meeuse (1942) stated:
The latex cells thereforedo not slide on the walls of their neighborcells. The
relativedisplacementof the tips of the young latex cells in the embryotherefore
probablycomes about in the same way as in the case of crystalcells of Citrus,
i.e., the cells alter theirrelativepositionchieflyby symplasticreadjustmentof the
wholeframeworkof cell walls, . . .
It was not knownhow the variousgrowthprocessesof the laticiferwerecontrolled.
Tip growth,cell elongation,rateof growth,directionof growth,formationof branches
and redifferentiationin the laticiferrepresentedcomplex processesin cellulardifferentiation.Since Schaffstein'sinvestigations(1932) were publishedshortlyafterWent
(1928) introducedhis conceptof how auxinaffectedplantgrowth,it was not surprising
that Schaffsteinattributedthe various laticifergrowthphenomenato a chemotropic
Schaffsteinsuggestedthat auxin produced in the apical meristem formed a decreasing gradient as it diffused downwardlyalong the axis. He suggestedthat if
laticifers were sensitive to auxin they would grow along the gradient in the axis
towardthe meristem. Similarly,an auxin gradientextendingfrom the leaf primordium into the shoot might induce some of the branchesproducedby the laticiferto
grow into the primordium.

VIII. Wall Structureof the NonarticulatedLaticifer

The laticifer formed a soft, plastic primary wall during its early development
(Sperlich, 1939). However, the thickness of this wall was quite variable. In some
plants, such as Urtica,it was very delicate, perhapsonly as thick as the wall of the
adjacentparenchymatouscells. On the other hand, in some species of Euphorbia,as
E. splendens,the wall could attain a thickness of 10-16 ,um(Schwendener,1885).
The chemical and physical structureof the laticiferwall in E. splendenswas investigatedby Frey-Wyssling(1926, 1932). Upon differentialdissolution of cell wall
constituents,he reportedthat this wall was relativelyhigh in pectic materialto which
he attributedits hygroscopiccharacter.
Frey-Wyssling(1942) investigated the micellar arrangementin the walls of the
laticifer.Wall birefringenceand iodine dichroism indicated that cellulosic micellae
were somewhat randomlyarranged.In a transectionalview of the cell the micellae
were arrangedtangentiallyto the wall surface,whereasin longisectionthe cellulosic
micellae were arrangedpredominantlyin the horizontalplane. He termed this particular arrangementthe tube structure.It was not peculiarto the laticifer,but also
characterizedwalls of sieve tubes, elongated parenchymacells, meristematiccells,
as well as the primarywalls of bast fibers and all cambium derivatives (Meeuse,
The wall of a laticiferwas often very irregularlythickened.This was most evident
in those plants in which the laticiferwall was characteristicallythickerthan that of
the adjacentcells (Euphorbiasp.). Irregularityin thicknessof the cell wall appeared
to be a result of the plasticity of the wall. Hence, as the tip of the laticifercell grew
along the intercellularspace it conformedto the contoursof the adjacentcells.



The presenceof primarypit-fieldsin the wallsof laticiferouscells has beendescribed

only by a few investigators.Haberlandt(1883) contended that he observed pits in
the laticifersof Euphorbialathyriswherelaticiferscontactedpalisadecells, and more
obvious pits were evident where they contacted spongy parenchymacells. Protoplasmic connectionsbetween a laticiferand the adjacentparenchymatouscells were
reportedin Neriumoleander(Haberlandt,1886; Kienitz-Gerloff,1891; Strassburger,
1901; Trecul, 1866), in Euphorbiacyparissias(Kienitz-Gerloff,1891), in Ceropegia
(Borscow,1869) and also in CodiaeumvariegatumBl. (Bottcher,1927). Reess (1896)
reportedlyobserved primary pit-fields in the wall of laticifers, but was unable to
detect the correspondinghalf of the pit-field in the adjacentcell.
IX. Cytology of the Nonarticulated Laticifer
The cytology of nonarticulatedlaticifershad received little attention in classical
anatomicalstudies of the cell. Even De Bary'sreview (1884) on the topic of laticifers
was noticeably lacking in a discussion of any cytological aspects. This quotation
expressesthe state of knowledgeof the microscopiccontentof these cells at that time:
Withinthe wall neitherprotoplasmnor nuclei are to be seen. It is true many
forms of coagulated,finely granularlatex, as that of the Cichoriaceae,resemble
coagulatedprotoplasm,or theirremainshereand there,in partiallyemptiedtubes
afteractionof alcohol,solutionof iodine,etc., a coat whichlookslike a coagulated
protoplasmiclining to the wall. Furtherinvestigationswill thereforeperhapsbe
able to prove the presenceof a protoplasmicbody.
The uncertaintyof the protoplasmiccontent within laticifersresultedin such interpretationsas advancedby Berthold(1886), who temporarilyresolvedthe problem
by referringto the entire content as protoplasm.However, Schmidt (1880), Kallen
(1882) and Molisch(1901) wereableto distinguisha marginalprotoplasmin laticifers
of Euphorbiasplendens,EuphorbiapulcherrimaWilld. and Ficus elastica. The protoplasmic lining was not sharply distinguishablefrom the vacuolar content, but
appearedto intergradewith it. From their investigationsupon living tubes of Euphorbia, Bobilioff (1925) and Popovici (1926) described laticifers as possessing a
marginalprotoplasmicsheathwhich surroundeda largecentralvacuole. The cell sap
of the vacuole consisted of various inorganicsubstancesas well as starchand other
Presenceof nuclei within laticiferswas observed by several workers(Buscalioni,
1898; Haberlandt,1883; Molisch, 1901; Nemec, 1910; Smolak, 1904; Treub, 1879,
1880). Treubidentifiedthe laticiferas multinucleated,which induced Foster (1949)
to describe the nonarticulatedlaticifer as a coenocyte. Treub, the first to describe
laticifer cytology, depicted the nuclei as large and peculiarly spindle-shaped.He
observed them in various genera including the Urticaceae, Apocynaceae,Euphorbiaceae, and Asclepiadaceae.The uniformity in appearanceof the nuclei and the
regularityof their distributionwithin the laticifersuggestedto Treub that they were
all in a similar state of development and activity. Occasionallyhe observed several
nuclei present in a group, but usually they were quite uniformly spaced along the
Originof the multinucleatedprotoplastin the laticiferwas ascertainedonly with
difficulty.The most logical assumption was that the coenocytic condition resulted
from repeatednucleardivisions without the formationof cell walls. However, most



of the investigatorsapparentlywere unable to detect the occurrenceof any nuclear

divisions withinlaticifers.Onlya few observationsof karyokinesishave been reported
(Buscalioni,1898;Nemec, 1910;Treub, 1880). Treubobservedand depictednuclear
division figuresin laticifersof UrticadioicaL., VincaminorL., Stephanotisfloribunda
Brogn.,Hoya ariadneDecne., OchrosiacoccineaMiq. and Cyrtosiphoniaspectabilis
Miq. Karyokinesis,as he describedit, was a rhythmic process in which the nuclei
divided simultaneously.Althoughhe implied that all nuclei along the entire length
of the tube performedin this manner,he did not state this explicitly.Nemec (1910)
could not substantiatethe occurrenceof simultaneousnucleardivisions in these cells.
Rather,he observedkaryokineticactivityto be restrictedto the plantmeristemwhere
adjacentparenchymatouscells were also dividing. No nuclear divisions were seen
in the laticiferbelow the meristem in the region of cell elongation.
X. Summary
The classical studies of laticifers have contributed seminal information on the
structureand development of these cell types in vascularplants, and provided the
bases for additionalinvestigationsinto their organizationand interrelationships.The
association of unusual cell contents with laticiferscontributedto their being interpretedas secretorystructures(Esau, 1953).This designationprovideda workingbasis
for studies on these cells, and the results and implications from the studies of the
classical workershave raised major questions on the characterand relationshipsof
these cells.
The implications of the secretorydesignation suggest a functional role for cell
contents or for the cell itself. There are few data to support a functional role for
laticifersalthoughone can speculatethat these specializedcells do performfunctions
in the plant. Indeed, laticifers in differentgenera and families may have evolved
differentfunctions,with no one function characterizinglaticifersin general.
The recognitionof two distinctivelaticifertypes, the articulatedand nonarticulated
laticifers,has provideda workinghypothesisfor examiningtheir ontogeny.Although
this classificationhas been convenient, it may be too simple and, therefore,obscure
subtledifferenceswithin each or both types that warranttheir furtherseparationinto
The recognitionof differentontogeneticorigins of laticifersfor plants in different
families brings into question the homology between nonarticulatedand articulated
laticifers.Their assumedhomology is typicallybased on the natureof their "milky"
contents ratherthan on patternsof differentiationand ontogeny. Cell contents, as
already shown by classical workers,can vary between differentlaticifers and may
reflectselective processesindependentfrom those controllingcell differentiationand
ontogeny. The nonarticulatedlaticifer may be a distinctive cell type whereas the
articulatedlaticifermay be relatedphylogeneticallyto other idioblasts.
The disjunct distributionof each laticifertype among only a few families of angiospermsas noted by the classicalworkerswould indicate a polyphyleticorigin for
both articulatedand nonarticulatedlaticifers.Althoughpast studiesindicatea limited
distributionin angiosperms,future studies may reveal a more wide-spreaddistribution than presentlyrecognized.Their absence in primitive angiospermssuggests
that these cells are more recent in origin than most other cell types.
The greatlengthof the nonarticulatedlaticifer,it being the longestof all biological
cells, contributedto an interpretationthat the mechanism which limits the growth



in length of other cells is absent in this laticifer.Thus, unlike other cell types, this
laticifercan grow intrusivelyas a single cell throughoutdifferenttissues of the plant
body. This featuresets the nonarticulatedlaticiferapartfrom other cells.
The detectionby classicalworkersof a coenocyticprotoplastin the nonarticulated
laticiferemphasizedthe occurrenceof an alterationin the mitotic apparatusresulting
in the loss of the mechanismcontrollingcell plate and wall development.All nuclei
in a nonarticulatedlaticifer, including those in the growing tips of the laticifer in
plant meristems, representprogenyfrom the originalnucleus of the laticiferinitial.
This laticifer,therefore,is cytologicallyisolatedfrom the meristemswhereasfor other
tissues the meristemscontributenew cells with their nuclei to the developingtissue.

XI. Acknowledgments
The authorwishes to thank Dr. SigridLiede for preparingthe Germantranslation
of the abstractfor this review.

XII. LiteratureCited
Anonymous. 1846. Die Milchsaftgefasse,ihr Ursprungund ihre Entwicklung.Bot. Zeitung4:
833-843, 849-859, 865-872.
Bernhardi,J. J. 1805. Beobachtungenilber Pflanzengefasseund eine neue Art derselben.In
der Hennings'chen Buchhandlung,Erfurt.
Berthold,G. 1886. Studieniuberprotoplasmamechanik.
A. Felix, Leipzig.
Blaser, H. S. 1945. Anatomy of CryptostegiagrandifloraR. Br. with specialreferenceto the
latex system. Amer. J. Bot. 32: 135-141.
Bobilioff,W. 1925. Observationson latex vessels in the living condition. Arch. Rubbercult.
9: 313-342.
Bonner,J. & A. W. Galston. 1947. The physiologyand biochemistryof rubberformationin
plants. Bot. Rev. 13: 543-596.
Borscow, E. 1869. Ueber gegitterteparenchymazellein Rinde et Stengels von Ceropegia
aphyllaund deren Beziehungzu der Milchaftgefassen.Jahrb.Wiss. Bot. 7: 344-356.
Bottcher,0. 1927. Die anatomisch-physiologischen
Beziehungender Milchr6hrenzum Assimilationssystem.Beih. Bot. Centralbl.44: 218-240.
Buscalioni,L. 1898. Osservazionie richerchesullacellulavegetale.Ann. RealeIst. Bot. Roma
7: 255-346.
Calvert, A. 1887. The laticiferoustissue in the stem of Hevea brasiliensis.Ann. Bot. 1:
Cameron,D. 1936. An investigation of the latex systems in Euphorbiamarginata,with
particularattention to the distributionof latex in the embryo. Trans. Proc. Bot. Soc.
Edinburgh32: 187-194.
Chandler,L. 1933. OxfordEnglishdictionary,vol. VI. ClarendonPress, London.
Chauveaud,M. G. 1891. Recherchesembryogeniquessur l'appareillaticiferedes Euphorbiacees, Urticacees,Apocyneeset Asclepiadees.Ann. Sci. Nat. Bot., ser. 7, 14: 1-161.
David, G. 1872. Uber die Milchzellender Euphorbiaceen,Moreen, Apocyneenund Asclepiadeen.Dissertation,Breslau.
De Bary, A. 1884. Comparativeanatomy of the vegetative organsof the phanerogamsand
fems. Englishtransl.by F. 0. Bowerand D. H. Scott. ClarendonPress, Oxford.
Dippel, L. 1865. Entstehungder Milchsaftgefasseund deren Stellungin dem Gefassbundelsysteme der milchendenGewiichse.Dissertation,Rotterdam.
Esau, K. 1953. Plant anatomy.John Wiley and Sons, New York.
Faivre, E. 1866. Recherchessur le circulationet sur le role au latex dans le Ficus elastica.
Ann. Sci. Nat. Bot., ser. 5, 6: 31-51.



1868. Etude sur le latex du murierblanc. Ann. Sci. Nat. Bot., ser. 5, 10: 97-122.
Foster,A. S. 1949. Practicalplant anatomy,ed. 2. D. Van Nostrand,New York.
Fraser, L. 1931. An investigationof Lobelia gibbosa and L. dentata. I. Mycorrhiza,latex
system and generalbiology. Proc. Linn. Soc. N.S.W. 56: 497-525.
Frey-Wyssling,A. 1926. Die submicroskopischestrukturder Zellenmembranen.EinepolarizationscoptischeMethodesum Nachweisder Richtigkeitder Mizellartheorie.Jahrb.Wiss.
Bot. 65: 195-223.
1932. Der Milchsaftergussvon Heveabrasiliensisals Blutungserscheinung.
Ein Beitrag
zur Druckstromtheorie.Jahrb.Wiss. Bot. 77: 560-626.
1942. Uber Zellwiindemit R6hrentextur.Jahrb.Wiss. Bot. 90: 705-730.
Ann. Sci. Nat. Bot., ser.
Gaucher,L. 1902. Recherchesanatomiquessur les Euphorbiac&es.
8, 15: 161-310.
Gravis,A. 1884. Recherchesanatomiquessur les organesv6getatifsde 1'UrticadioicaL. Extr.
Mem. Acad. Roy. Belgique47: 1-256.
Grew, N. 1682. Anatomy of plants, with the idea of a philosophicalhistory of plants. W.
Rawlins, London.
Groom,P. 1889. On the function of laticiferoustubes. Ann. Bot. 3: 157-165.
Guerin,P. 1923. Les Urticees:CellulesA mucilage,laticifereset canauxsecreteurs.Bull. Soc.
Bot. France70: 125-136, 207-215, 255-263.
Haberlandt,G. 1883. Zur physiologischenAnatomie der Milchrohren.Sitzungb. Kaiserl.
Akad. Wiss., Math.-Naturwiss.KI., Wien 87: 51-69.
1886. PhysiologischePflanzenanatomie
et uiberdasAssimilationsystemen.
Bot. Ges. 4: 206-236.
1914. Physiologicalplant anatomy,4th. ed. Englishtransl.by M. Drummond.Macmillan, London.
Hanstein,J. 1864. Die Milchsaftgefasseund die verwandtenOrganeder Rinde. Berlin.
Hartig, T. 1843. Beitriagezer Entwickelungsgeschichte
der Pflanzen.A. F6rstner,Berlin.
. 1862. Uber die Bewegungde Saftesin der Holzpflanzen.Bot. Zeitung.20: 73-76, 8187, 89-94, 97-100, 105-109.
Harvey, W. 1941. Exercitatioanatomicade motu cordis et sanguinisin animalibus,ed. 3.
Englishtransl.by C. D. Leake.C. Thomas, Baltimore.
Hooke, R. 1665. Micrographia.Martinand Allestry,London.
Jackson, B. D. 1928. A glossaryof botanic terms, ed. 4. Ed. Duckworth,London.
Kallen,F. 1882. Verhaltendes protoplasmasin den Gewebenvon Urticaurens.Flora65: 6880, 81-92, 97-105.
Kienitz-Gerloff,F. 1891. Die Protoplasmverbindungzwischen benschbartenGewebeselementen im der Pflanze.Bot. Zeitung49: 17-26.
Lindley, John. 1848. Introductionto botany, ed. 3. Longman, Orne, Brown, Green and
Link, H. 1824. Elementaphilosophiaebotanicae.Haude et Spener,Berolini.
1837. Grundlehrender Krauterkunde.Haude und Spener,Berlin.
Malpighi, M. 1901. Die anatomie der Pflanzen, 1675-1679, 2 parts.Germantransl.by M.
Mobius. W. Engelmann,Leipzig.
Mariotte,E. 1717. Oeuvresde Mr. Mariotte.Jean Neaulme, Le Havre.
Mayus, 0. 1905. Beitriigeiiber den Verlauf der Milchr6hrenin den Blattem. Beih. Bot.
Centralb.18: 273-286.
Meeuse, A. D. J. 1942. A study of intercellularrelationshipsamong vegetable cells with
special referenceto "slidinggrowth"and to cell shape. Recueil Trav. Bot. Neerl. 38: 18140.
Meyen, F. J. F. 1838. Neues System der Pflanzenphysiologie,3 vols. Haude und Spener,
Mirbel, C. 1815. Elementsde physiologyvegetaleet de botanique,2 vols. Maginel,Paris.
Mohl, H. von. 1844. EinigeBermerkungenuberden BaudervegetabilischenZell.Bot. Zeitung
2: 273-277, 289-294.



* 1852. Principlesof the anatomyand physiologyof the vegetablecell. Englishtransl.

by A. Henfrey.J. van Voorst, London.
Moldenhauer,J. H. D. 1812. Beitragesur anatomieder Pflanzen.K6nig. Schulbiicherei,Kiel.
Molisch, H. 1901. Studieniuberden Milchsaftund Schleimsaftder Pflanzen.G. Fischer,Jena.
Nemec, B. 1910. Das problem der Befruchtungsvorgiinge
und andere zytologischeFragen.
Parkin,J. 1900. Observationson latex and its functions.Ann. Bot. 14: 193-214.
Pax, F. 1884. Die anatomie der Euphorbiaceenin ihrer Beziehungzum system derselben.
Engler,Bot. Jahrb.5: 384-421.
Pitra, A. 1860. Ueber das Verhaltnissder Milchsaftgefassesu den Bastzellen.Bull. Soc. Imp.
Nat. Moscou 33: 203-222.
Popovici,H. 1926. ContributionsA l'etude cytologiquedes laticif'eres.Compt. Rend. Hebd.
SeancesAcad. Sci. 183: 143-145.
Potter, M. C. 1884. On the development of starch grains in the lactiferouscells of the
Euphorbiaceae.J. Linn. Soc. 20: 446-450.
Reess, M. 1896. Lehrbuchder Botanik.F. Enke, Stuttgart.
Schacht,H. 1851. Die sogenanntenMilchsaftgefasseder Euphorbiaceen.Bot. Zeitung9: 513521.
. 1856. Uber die Milchsaftegefisseder Carica Papaya deren Entstehung,Bau und
Verlauf.Monatsber.Akad. Wissensch.,Berlin, 515-534.
Schaffstein,G. 1932. Untersuchungenan ungegliedertenMilchr6hren.Beih. Bot. Centralbl.
1, 49: 197-220.
Schleiden,M. J. 1844. Beitragezur botanik.W. Engelmann,Leipzig.
1849. Grundziugeder wissenschaftlichenBotanik,Vol. 1. W. Engelmann,Leipzig.
Schmalhausen,J. 1877. Beitragezur Kenntnisder Milchsaftbehalter.MWm.Acad. Imp. Sci.,
St. Peersb., ser. 7, 24: 1-27.
Schmidt,E. 1880. Etudecompar6edes ecores de tiges et de racinesde quelquesEuphorbes.
Schullerus,F. 1882. Die physiologischeBedeutungder Milchsaftesvon EuphorbiaLathyris.
Verh. Bot. VereinsProv. Brandenberg24: 28-93.
Schultz, C. H. 1839. Sur la circulationet sur les vaisseaux laticiferesdans les Plantes. A.
1841. Die Cyclosedes Lebenssaftesin den Pflanzen.Verh.Kais.Leopold-Carol.Akad.
Naturforsch.18 (suppl. 2): 1-355.
Schwann,T. 1839. MikroskopischeUntersuchungeniber die Uebereinstimmungin derStruktur und dem Wachsthumder Thiere und Pflanzen.Sander'schenBuchhandlung,Berlin.
Schwendener,S. 1885. EinigeBeobachtungenan Milchsaftgefassen.Sitzungsb.Akad. Wiss.,
Berlin 1: 332-337.
Scott, D. H. 1884. On the laticiferoustissue of Manihotglaziovii.Quart.Jr. Microscop.Sci.
(N. S.) 24: 194-204.
Smolak,J. 1904. Uber vielkemigeZellenbei einigenEuphorbiaceen.Rozp. CeskeAkad.Cis.,
Bull. Intern.9: 135-149.
Solereder,H. 1908. Systematicanatomyof the dicotyledons.Englishtransl.by L. A. Boodle
& F. E. Fritsch.ClarendonPress, Oxford.
Sperlich,A. 1939. Das trophischeParenchym.B. Exkretionsgewebe.
In K. Linsbauer,Handbuch d. Pflanzenanatomie,Bd. 9. GebruderBortraeger,Berlin.
Sprengel,K. 1817. Anleitungzur Kenntnisder gewachse.K. A. Kummel, Halle.
Spry, E. 1767. Account of a lockedjaw, and paralysiscuredby electricity.Philos. Trans.57:
Strassburger,E. 1901. Uber PlasmaverbindungenPflanzlicher.Jahrb.Wiss. Bot. 36: 493610.
Theophrastus. 1916. Enquiryinto plants. English transl. by Sir A. Hort. W. Heinemann,



Trecul,A. 1860. Rapportdes laticiferesavec le systemefibro-vasculaire.Compt.Rend. Hebd.

SeancesAcad. Sci. 51: 871-874.
1865a. Matiereamylaceeet cryptogamesamyliferesdans vaisseauxdu latex de plusieursApocynees.Compt. Rend. Hebd. SeancesAcad. Sci. 61: 156-160.
1865b. Laticifereset liber dos Apocyneeset des Asclepiadees.Compt. Rend. Hebd.
SeancesAcad. Sci. 61: 294-298.
1865c. Surles laticifereset les fibresdu liberramifieesdans les de Euphorbes.Compt.
Rend. Hebd. SeancesAcad. Sci. 60: 1349-1352.
1866. Resume d'observationssur les vaisseaux et les sucs propres.Ann. Sci. Nat.
Bot., ser. 5, 5: 44-79.
Treub,M. 1879. Sur la pluralitedes noyaux dans certainescellulesvegetales.Compt. Rend.
Hebd. SeancesAcad. Sci. 89: 494-496.
1880. Sur des cellules vegetalesa plusieursnoyaux. Arch. NMerl.Sci. Exact.Nat. 15:
Treviranus,L. C. 1835. Physiologieder Gewachse.Bonn.
Tschirch,A. 1889. AngewandtePflanzenanatomie,2 vol. Urban und Schwarzenberg,Wein.
Unger, F. J. 1840. Beitragezur Kenntnis der parasitischenPflanzen. Ann. Wiener Mus.
Naturgesch.2: 13-60.
1846. Grundziugeder Anatomie und Physiologieder Pflanzen.C. Gerold,Wien.
1847. Die Intercellularsubstanz
und ihr Verhaltnisszur Zellmembranbei Pflanzen.
Bot. Zeitung5: 289-300.
1858. Beitriigezur Anatomie und Physiologieder Pflanzen.Sitzungsb.Akad. Wiss.,
Math.-Naturwiss.KI., Wien 12: 367-396.
Vogl, A. 1863. Uber de intercellularsubstanzund die Milchsaftgefassein der Wurzel des
gemeinenL6wenzahns.Sitzungsb.Akad. Wiss., Math.-Naturwiss.K1.,Wien 48: 668-690.
1866. Beitriagezur Kenntnis dcs Milchsaftorganeder Pflanzen.Jahrb.Wiss. Bot. 5:
Vreede, M. C. 1949. Topographyof the laticiferoussystem in the genus Ficus. Ann. Bot.
Gard. Buitenzorg51: 125-149.
Went, F. W. 1928. WuchsstoffundWachstum.Recueil Trav. Bot. NMerl.25: 1-116.
Wolff, C. F. 1869. Theoriagenerationis(1759). Germantransl.by P. Samassa.Engelmann,

Zander,A. 1928. Uber Verlaufund Enstehungder Milchrohrendes Hanfes(Cannabissativa).

Flora 23: 191-218.