A Criticism of the Cell-Theory; being an Answer to Mr ......A OEITIOISM OF THE GULL-THEORY. 137 A...

A OEITIOISM OF THE GULL-THEORY. 137

A Criticism of the Cell-Theory; being an Answerto Mr. Sedgwick's Article on the Inadequacyof the Cellular Theory of Development.

By

Gilbert C. Bourne, M.A., F.L.S.,Fellow of New College, Oxford.

"Jedes Lebendige isb kein Einzelnes, sondern ein Melirheit; selbst inso-fern es uns als Individuum erscheint, bleibt es doch eine Yersammlung youlebendigen, selbstandigen Wesen, die der Idee, der Anlage nach gleich sind,in der Erscheinung aber gleich oder ahnlich, ungleich oder unahnlich werdeaIconncn. Diese Wesen sind theits urspriinglich schon verbunden, theils findenund vereinigen sie sich. Sie entzweien sicb und suclien sich wieder, undbewirken so eine unendliche Production auf alle Weise und nach alienSeiten."—GOETHE (1807).

MR. ADAM SEDGWICK has of late thrown himself withconsiderable zeal into the part of a zoological iconoclast, andhas displayed an evident relish in battering the idols which,he would fain make us believe, are turning away the minds ofmen from the true faith, of which there are but few orthodoxexponents. Nor may we blame him for his fervour, for an oldfaith always emerges purer, if not firmer, from the ordeal ofsharp antagonism. The idols in question are the develop-mental law of von Baer and the cell-theory.

Seeing how important a thing it is that a science should beguided by principles capable of being expressed in preciselanguage, it has been a matter of surprise to me that somecompetent person has not taken up the challenges whichMr. Sedgwick has thrown down. For, if his views are topf evail, two of the fundamental principles of zoology, principleswhich have hitherto directed and steadied the course of zoolo-gical speculation, are taken away from us; and unless some

138 . GILBERT 0. BOURNE.

better and more distinct principles are put in their place, thecourse of speculation may be expected to be very erraticindeed. It is not without serious misgivings as to my owncompetence that I, in default of a better champion, take upone of these challenges, and I propose to criticise Mr. Sedg-wick's recent article on the inadequacy of the cellular theoryof development, leaving for a future occasion the considerationof his earlier article on von Baer's law.

It is to be regretted that Mr. Sedgwick should, in puttingforward a view aflFecting one of the fundamental propositionsof biology, have chosen to adopt a controversial method, whichcannot but have the effect of weakening his case. And it isstill more a pity that he should be so unsparing in abuse ofhis imaginary opponents, whilst he himself commits the veryfault for which he so much blames them. For he lays, in thefront of his indictment, a charge of vagueness and unsub-stantiality against the supporters of the cellular theory." We are dealing," he says, " with a kind of phantom whichtakes different forms in different men's eyes. There is a wantof precision about the cell-phantom, as there is also about thelayer-phantom, which makes it very difficult to lay either ofthem. Neither of these theories can be stated in a mannersatisfactory to every one. The result is that it is not easy tobring either of them to book."

I shall show, later on, that this charge of vagueness is notaltogether justified; what I am at present concerned with isto show that Mr. Sedgwick is as much open to the charge ofvagueness as the rest of the zoological world which he casti-gates.

Read his article through as carefully as one may, onecannot find any definite or precise statement of his own stand-point, saving that he quotes passages from one of his earlierworks. The critic, therefore, mnst be content to inferfrom the tenor of the whole article, and from particularpassages in it, as well as from his previous writings, whatMr. Sedgwick does or does not believe with regard to the cell-theory, and if he is misinterpreted, it is his own fault.

A CRITICISM OP THE DELL-THEORY. 139

It is probably a fair summary of his position to say that, forthe present, he limits his objections to the application of thecell-theory to the process of growth during embryonic deve-lopment j but that he scarcely conceals his preference for theview that there are no such things as discrete cells in theso-called multicellular organism. And as it is necessary, atthe outset, to have a perfectly clear idea of his meaning, I willquote passages from the work to which he refers in his openingparagraph, assuming that what he stated then he is preparedto adhere to now, and that his last article is intended toemphasise the views which he formerly propounded, and tobring fresh evidence in support of them.

On p. 204 of the second part of his account of the deve-lopment of the Cape species of Peripatus, he says:—"It isbecoming more and more clear every day that the cells com-posing animal tissues are not isolated units, but that they areconnected with one another. I need only refer to the connec-tion known to exist between connective tissue cells, cartilagecells, epithelial cells, &c. And not only may the cells of onetissue be continuous with one another, but they may also becontinuous with the cells of other tissues. . . . It is truethat the cells of blood and lymph and the ripe generative cellsare completely isolated. But the former, in their first stagesof growth, form part of the syncytium, as in all probability dothe latter also. This continuity, which for a priori reasonswe should expect, has hitherto been regarded as a fact of littlemorphological importance and relegated to the category ofsecondary features. The ovum, it is said, segments intocompletely isolated cells, and the connection between them isa secondary feature acquired late in development. It hasalways been considered that the first stage in the evolutionof the Metazoa was a colonial Protozoon, i. e. a mass of perfectlyisolated unicellular organisms, derived by complete divisionfrom a single cell. Now while I do not wish to exalt the factsof the cleavage and early development of Peripatus to a posi-tion of undue importance, or to maintain that of themselvesthey are sufficient to destroy this conception of the origin and

140 GILBERT 0. BOUENB.

structure of a Metazoon, I think I am justified in pointingout that, if they are found to be of general application, ourideas on these subjects trill have to undergo considerablemodification. The ancestral metazoon will no longer belooked upon as a colonial protozoon, but rather as havingthe nature of a multinucleated infusorian, with a mouthleading into a central vacuolated tract of protoplasm. Thecontinuity between the various cells of the adult—the connec-tions between the nerves and muscles und sensory epithelium,receive an adequate morphological explanation, being due toa primitive continuity which has never been broken. Inshort, if these facts are generally applicable, development canno longer be looked upon as being essentially the formation ofa number of units from a single primitive unit, and theco-ordination and modification of these units into a harmoniouswhole. But it must rather be regarded as a multiplication ofnuclei and a specialisation of tracts and vacuoles in a con-tinuous mass of vacuolated protoplasm."

This is a temperate and lucid statement of a suggestionwhich is still worthy of serious consideration, the more sosince it had been shown, but a short time previous, thatprotoplasmic continuity between the tissue-cells of plants is ofvery general occurrence, if not the rule. And, as a historicalfact, the continuity of protoplasm was a phenomenon familiarto animal histologists long before it was proved for vegetabletissues; indeed there were authors who, before Mr. WalterGardiner's researches were published, were disposed to regardprotoplasmic continuity as a characteristic of animal organisa-tion, discontinuity as a characteristic of vegetable organisa-tion.

I have quoted at length because Mr. Sedgwick from beingtemperate has become intemperate, and from being lucid hehas become obscure; so that, were I to deal only with hislatest utterances, I should be quite at a loss to know what hismaturer views might be.

What follows, then, may be taken to be a not unfair state-ment of his position. That from the connection known to

A CRITICISM OP THE OELL-THEOKY. . 141

exist between some cells composing adult tissues, there is anantecedent probability that similar connections exist betweenall cells composing all tissues; and this probability is heightenedby observations made on the development of Peripatus, by thefact that the so-called mesenchyme cells in Avian and Selachianembryos are continuous, and not isolated, as was once supposed,and by a study of the developing nerves of Elasmobranchs.And that it follows from this that the morphological conceptof a cell, so far from being of primary, is altogether of Secon-dary importance, and that progress in the knowledge ofstructure is impossible so long as men persistently regardcells as the fundamental structural units on which the pheno-mena manifested by organised beings depend. The truemethod of enquiry must be a study of the growth, extension,vacuolation and specialisation of the living substance—proto-plasm.

It is in this sense that I propose to deal with Mr. Sedgwick'sviews, and he will pardon me if I have misinterpreted them.At any rate, I have done my best to understand them.

I would wish to show, in the first place, that there is veryslender ground for the accusations which Mr. Sedgwick levels,in an unsparing manner, against his zoological contemporaries.He goes so far as to say that their eyes are blinded by theoryto the most patent facts, and that " they are constrained bythis theory,"—the cell theory,—" with which their minds aresaturated, not only to see things which do not exist, butactually to figure them." This is abuse and not argument;if Mr. Sedgwick were to remember the qualifying sentence inhis writings of 1886, " i f they are of general application," hewould recognise that there is little occasion for accusingZoologists of perversely ignoring the views which he thenSet forth.

For, in fact, the phenomena to which he draws our attentionhave received their due meed of recognition from the timethat the cellular structure of tissues was first studied,More recent researches have enlarged our knowledge of proto-plasmic continuity, but it is still a phenomenon far from being

142 GILBERT 0. BOURNE.

of such universal application as to constrain us to abandonthat very useful morphological concept—a cell.

For some years past the study of cells, of their ultimatestructure, of their chemical and physical properties, of pheno-mena which accompany their growth and division, has beencarried on with a minuteness which a short time ago wasundreamt of. And attention has been directed, not only tothe cells composing adult tissues, but in the most markeddegree to the successive formation of cells from the primitiveunit, the oosperm, and to the fate which each subsequentlyundergoes in the course of development. In place of the off-hand statements of older embryologists, that the ovum dividesinto two, four, eight, sixteen segments, and so forth, we havethe most accurate and minute accounts of the successiveformation of cells, of the place which each occupies in thedeveloping embryo, of its parentage and of its progeny, andof the share taken by the last named in the building up of theadult tissues. In short, we have a number of cell-lineages,which show that in a number of animals, some of which arewidely separate from one another, the formation of cells fromthe ovum follows courses which are either identical or soclosely similar that the differences excite our wonder far leBSthan the similarities. So minute are these investigations thatevery karyokinetic figure has been followed in every cell, upto a stage where their number becomes bewildering.

I refer, of course, to the remarkable series of observationswhich were begun by Selenka, Arnold Lang, Hallez, Bloch-mann, and others, and have been carried to the highestperfection by von Wistinghausen, E. B. Wilson, Heymons,and Lillie.

It would be impossible, in such an essay as this, to dealadequately with the results obtained by these authors; and itis unnecessary, since their works are within reach of everyone.It is enough to say here that a perusal of them does not tendto diminish the importance which we have been accustomed toattribute to the cell in developmental processes.

Nothing can be more clear than the fact that, in Nereis or

A CRITICISM OF THE CELL-THEORY. 143

in Unio, there result from the division of the ovum separateprotoplasmic corpuscles, as distinct from one another as oneroom in a house is distinct from another, each of which is notonly separate, but contains within itself definite, and probablylimited, qualities (at least at stages beyond eight or sixteencells). One might almost say that, after the earliest stages,each blastomere has a definite task allotted to it, which itfaithfully and punctually performs, according to a prescribedcourse. To each, it might be said in figurative language, is-given material, which it must place, not anywhere, but in oneparticular part of the edifice.

In considering these very remarkable researches, it is notsufficient, for the present purpose, to say that no connectionbetween the blastomeres was observed. Such connectionsmay have existed and have been overlooked; as the con-nections, which undoubtedly exist, between plant cells werefor a long time overlooked. But, a pr ior i , such connectionsare improbable. For, as has been said, the qualities of eachblastomere are limited. Each is specialised before any formchanges become visible; each plays one part, and one partOnly in tissue formation. If their protoplasm were continuous,being made so by uniting strands, then, as Mr. Sedgwick hasexpressed it, the molecular constitution of any part would intime spread through the whole mass. But the molecularconstitution of the blastomeres must be different, for theirmanifestations are different, and we may possibly see, in thiscase, some explanation, obscure though it may be, of the iso-lation of the form elements from one another.

Further than this, there is objective proof that the cellsconstituting the early embryos of these forms are separate.They exhibit remarkable shiftings of position, which renderthe existence of connecting strands of protoplasm highly im-probable, and the migrations of some cells—e. g. those inNereis named c1>5< and d1>6> by Wilson—are of such an extentthat, if there were protoplasmic continuity, they would beimpossible.

It is no exaggeration to say that this is evidence which-


effectually disposes of the idea that a syncytial theory ofanimal orgaaisation is of general application.

It does more than this, it shows that there are not a fewinstances in which cells possess a morphological and physiolo-gical significance greater than was at one time supposed.

There are numerous other cases in which, at an early stageof development, cells wander far from the position in whichthey originated, and become placed so far from the parentcells from which they sprung, that any idea of protoplasmiccontinuity is impossible. As examples I may mention: theouter layer cells of Cornacuspougige and Silicispongise, which, asMaas has shown, go through remarkable migrations; the me-soblast of Cal l iani ra biala ta , Beroe and Cydippe, as de-scribed by Metschnikoff, whose statements are confirmed byobservations made (but unfortunately not published) by Mr.Riches on Hormiphora p lumosa; the lower endodermcells of Discocoelis, Eurylepta, and Leptoplana, as describedby Lang, Hallez, and Selenka.

In short, the evidence is overwhelming, and it must betaken to be very clearly established that there are numerouscases in which there is not " a primitive continuity which hasnever been broken."

It is apparent, then, that morphologists have been amplyjustified in refusing to recognise Mr. Sedgwick's views as tothe syncytial nature of animals, and there is no justificationfor the strong language which he uses towards them onaccount of their refusal.

It is, on the other hand, quite possible that the frequency ofthe occurrence of protoplasmic continuity between developingtissue-cells may have been overlooked or ignored by a fewauthors, and that those who have done so have been led intothe error of attributing too great and too fundamental im-portance to the cell as an independent vital unit (Lebenseinheit).

But, in point of fact, I am unable to find, in the writings ofany reputable biologist, any statement to the effect that anorganism is composed of independent and isolated units. Onemay, it is true, find passages here and there which, when


removed from the context, might be made to bear such aninterpretation. I have questioned my pupils with regard tosuch passages, and I find that they do in fact put such aninterpretation upon them. For instance, in Waller's ' Intro-duction to Human Physiology' the following passage occurson page 2: " The organism is a community; its individualsare cells; groups of its individuals are organs." Here wehave an example of the danger of the too free use of illustrativelanguage. In every illustration there lurks a fallacy. Thefallacy may not have been present to the mind of the author;but if the illustration alone is used, without a lucid explanationof its meaning, the fallacy may be the one thing which im-presses itself on the minds of his readers. In this case thereis a fallacy in the analogy, so often made use of for purposes ofpopular exposition, between an organism and a community. Ifthe analogy is used without the necessary reservations it leadsto confusion, for the reader is only too prone to transfer tothe organic unit the idea of the individual isolated man, whois the social unit. The organic unit may in some cases beindividual and isolated, but in the great majority of instancesit has lost, wholly or partially, its individuality, and is notisolated. It becomes a subordinate part of a higher individality,which in its turn may be subordinate to an individuality of astill higher order. This has been explained in the most lucidand masterly manner by Hackel, in his ' Allgemeine Anatomieder Organismen,' published in 1866; and nobody who hascarefully studied that work can fail to have a clear under-standing of the subject. Yet it is to Hackel that the doctrineof a cell-republic is often attributed ! Clearly by those personsonly who have not read his works. For he insists, over andover again, upon a distinction (which since the researches ofMr. Walter Gardiner no longer holds good) between theorganisation of plants and that of animals, namely, that thespecial characteristic of plants lies in the preponderance of theperfected and differentiated individuals of the first order—thecells or plastids. "Der wesentliche tectologische Characterder Pflanzen liegt in der vorwiegenden Ausbildung und DifFer-

VOL. 3 8 , PAET 1.—NEW SEE. K

146 GILBERT 0. BOUBNE.

enzirung der Individuen erster Ordnung, der Plastiden" (op,cit.j p. 222). Of animals he says, on the contrary, " Derwesentliche tectologische character der Thiere liegt sowohl inder verwickelteren Zusammensetzung der Thierleibes aus weitdifferenzirten Individuen verschiedener Ordnung, als auchbesonders in der verschiedenartigsten Ausbildung der Indi-viduen zweiter Ordnung, der Organe, welche viel mannich-faltiger, als bei den Pflanzen und Protisten, differenzirt undpolymorph sind. Die Plastiden, die Individuen erster Ordnung,sind bei Thieren allermeist Zellen, und zwar meistens Nackt-zellen (ohne Membran) weniger Hautzellen (mit Membran).Sehr haufig, und allgemein in den entwickelten Personen,vereinigen sich bei den Thieren mehjere Nacktzellen zurBildung von Zellstocken (Nervefasern, Muskelfasem), was beiden Pflanzen nur bei der Bildung der Milchsaftgefasse undder Spiralgefasse geschiecht. Daher ver l ier t bei denThieren stets wenigstens ein Theil Zellen ihre in-dividuelle Selbs tandigkei t , wahrend sie dieselbein den Pflanzen meist behal ten ."

The last sentence, which I have put in italics, shows mostclearly that, as long ago as 1866, Hackel did not regard theanimal organism as a community, whose individuals are cells;and it is the fact that he applied the term " cell-republic " toplants, intending thereby to emphasise the difference whichhe believed to exist between vegetable and animal organisation.

So that, as a matter of history, whilst plants used to beconsidered to be colonies of independent life units, animalswere not. A certain exchange of opinion seems to have takenplace more recently. Some few zoologists and animal physi-ologists, borrowing from Hackel the term cell-republic, havethoughtlessly applied it, with all its implications, to animalorganisation, whilst botanists, influenced by Mr. Walter Gar-diner's researches, have insisted more and more upon theindividuality of the plant as a whole, and the subordination ofits component parts, the cells. None the less, the facts ofcell fusion and cell communication have never been whollyoverlooked by zoologists, and recent years have brought to

A CRITICISM OP THE CELL-THEORT. 147

light facts, such as the continuity of cartilage cells, which wereunsuspected when Hackel wrote.

I am therefore far from being satisfied that the independent-life-unit theory has had such a dominant influence as Mr.Sedgwick would have us believe; and I am quite certain thatthe picture which he draws of the teaching given to everystudent of biology is a travesty of the truth.

Biology includes botany as well as zoology, and if we wereto allow (which I do not) that zoologists generally havebecome as narrow in their conceptions of the processes ofdevelopment as Mr. Sedgwick says, it is quite certain thatbotanists have not. And as all students of biology are—or ifthey are not, they ought to be—put through a course ofelementary botany as well as of zoology (in many schools thesubjects are combined), grave blame must be imputed to thoseteachers who have, in the later stages of their education,Warped the liberal conceptions which they must have formedon the subject of organic growth and development. For I takeit that, after a study of Mucor, Vaucheria, and the Myxomy-cetes, there is no student so dull but he will have imbibedideas respecting cell growth which impel him to ask thequestion which as Mr. Sedgwick says it is so difficult to findan. answer to—" What, after all, is a cell?" If, when he asksthis question, he is told that the cell is an isolated corpuscleof protoplasm, the unit of vitality, and that there is " a mostfundamental distinction" between unicellular and multicellularorganisms, and so forth, the student may go on his wayrejoicing, for that he has at last been given a clear and tangiblestatement; but none the less he will have been started on avery wrong path. I have not a widespread experience ofzoological teaching, but I know, at least, that ProfessorLaokester's pupils are not started on that path. The truth is,and, if I am not much mistaken, zoologists and botanists alikehave long been possessed of it, that there is no fundamentalbut only a formal distinction between unicellular and multi-cellular organisms; that the cell is a form concept foundedon a very wide basis of experience, whereby we can conveniently


interpret to our minds one of the most universal of organicphenomena, viz. the splitting up of protoplasmic masses duringgrowth into a number of more or less distinct corpuscles.

It will not be out of place if I quote here a passage fromvon Sach's 'Vorlesungen iiber Pflanzenphysiologie5 (Englishedition, translated by H. Marshall Ward, 1887, p. 73). "Tomany the cell is always an independent living being, whichsometimes exists for itself alone, and sometimes becomes'joined with others'—millions of its like, in order to form acell colony, or as Hackel has named it for the plant parti-cularly, a cell republic. To others again, to whom the authorof this book also belongs, cell-formation is a phenomenon verygeneral, it is true, in organic life, but still only of secondarysignificance; at all events it is merely one of the numerousexpressions of the formative forces which reside in all matter,in the highest degree, however, in organic substance."

That this is a great limitation of the cell theory, both aspropounded by its authors and as held by many zoologists, isnot to be denied; and Mr. Sedgwick might well be contentif some such statement were made the established doctrineas regards cells. It appears to me that some such limitedstatement is necessary if we are to have any propositionuniversal ly applicable to organic structure; but with thisreservation, that I cannot regard as of secondary significancethat which all experience shows to be the expression parexcellence of organic growth.

In admitting this much, a large part of Mr. Sedgwick'sdemand is conceded, for it is not to be denied that the celltheory has been very differently and much more dogmaticallystated by quite recent authors.

We have, for instance, Dr. Oscar Hertwig's recent work,' Die Zelle und die Gewebe.' He begins dogmatically enoughby saying, "Thiere und Pflanzen, so verschiedenartig in ihrenausseren Erscheinung, stimmen in den Grundlagen ihres ana-tomischen Aufbaues iiberein; denn beide sind aus gleich-artigen, meist nur mikroskopisch wahrnehmbaren Elemen-tareinheiten zusammengesetzt. . . . Denn die Zellen, in


welche der Anatom die pflanzlichen und thierischen Organis-men zerlegt, sind die Trager der Lebensfunctionen, sie sind,wie Virchow sich ausgedriickt hat die 'Lebense inhe i t en . 'Von diesem Gesichtspunkt aus betrachtet, erscheint derGesammtlebensprocess eines zusammengesetzten Organismusnichts Anderes zu sein als das hochst verwickelte Resultatder einzelnen Lebensprocesse seiner zahlreichen, verschiedenfunctionirenden Zellen." The whole book is written "vondiesem Gesichtspunkt aus," and, admirable as it is, there isreason to think that its value is somewhat impaired by theexcessive value attributed to the cell as an independent vitalunit.

In passing, I may remark that this passage of 0 . Hertwig'sgives a very precise and definite statement of the cell theory,as it is held now, by a very great authority; and a referenceto older works would have shown Mr. Sedgwick that, so stated,it is practically the same as what its authors stated.1

For the original words of Schwann are these: " The ele-mentary parts of all tissues are formed of cells in an analo-gous though very diversified manner, so that it may be assertedthat there is one universal principle of development for theelementary parts of organisms, however different, and thatthis principle is the formation of cells In inferiorplants any given cell may be separated from the plant and cangrow alone. So that here are whole plants consisting of cellswhich can be positively proved to have independent vitality.Now, as all cells grow according to the same laws, and conse-quently the cause of growth cannot in one case lie in the celland in another in the whole organism, and since it may befurther proved that some cells, which do not differ from therest in their mode of growth, are developed independently,we must ascribe to all cells an independent vitality; that issuch combinations of molecules as occur in any single cell arecapable of setting free the power by which it is enabled totake up fresh molecules. The cause of nutrition and growth

1 ''lam not concerned with what its authors held."—Mr. Sedgwick, op. cit.,p. 88.

150 GILBERT 0. BOUBNE.

resides, not in the organism as a whole, but in the separateelementary parts, the cells."

The definitions of Hertwig are a re-statement in other wordsof the salient features of the theory of Schwann, and it is anerror to speak of an unsubstantial cell phantom. Nor is thereany unsubstantiality about the cellular theory of development,which, I may remind my readers, originated with Eemak.The cellular theory of development, taking as its starting pointthe conclusions of Schleiden and Schwann that all organismsare cells or composed of an aggregate of cells, states thatevery cell is formed by the division of a pre-existing cell, notas Schwann had supposed, by differentiation within a structure-less cytoblastema.1 Hence Virchow's well-known aphorism," omnis cellula e cellule," which, besides denying abiogenesis,expresses the cellular theory of development as succintly aspossible.

It would have been a great advantage to his own argument,and also to his critic, if Mr. Sedgwick had given the clear andauthoritative expositions of the cellular theory which lay readyto haud, instead of confusing the issue by a whimsical accountof his experience of morphological teaching.

Let us now examine the cell-theory, as stated by Hertwig,in the light of our present knowledge of animal and vegetablestructure.

It would not be a difficult task to demonstrate the generaltruth of Virchow's aphorism. Wherever there is a cell, it maybe shown to be the product, and generally the immediateproduct, of a pre-existing cell. But it would seem that somebiologists have added an unwarrantable corollary to Virchow'sgeneralisation, and would say, " Nil nisi cellula e cellula."Now from a certain aspect this might be considered true;everything depends on the question as to what is a cell?

Hertwig has pointed out, with much truth, that our pre-sent conception of a cell is inseparably connected with ourconception of protoplasm. We are still very far from under-

1 Mr. Sedgwick appears to have leanings towards a cytoblastema, as Ishall show further on.


standing the structure of protoplasm, and it might be saidthat, if we know nothing of the component, it is useless tomake assertions about the compost; but it will at least beuseful to criticise the attempts which have been made.

Hertwig gives this definition, which is the same as thatoriginally given by Max Schulze. A cell is a corpuscle ofprotoplasm in which is contained a specially organised con-stituent, the nucleus. (Die Zelle ist ein kliitnpchen von Proto-plasma, das in seinen Innern einen besonders geformtenBestandtheil, den Kern (Nucleus) einschliesst.) This at firstsight seems satisfactory enough, but the more one examinesit, the less satisfactory does it appear, in view of the differentkinds of organisms which are usually described as single cells.

If a corpuscle containing a nucleus is a cell, is a corpuscleContaining two or more nuclei also a cell ? And still more, isa large mass of protoplasm containing many nuclei to beregarded as a cell ? Such a mass, I mean, as Botrydium,Caulerpa, or Codium, or even Pelomyxa. By many authorsthese organisms are regarded as single multinucleate cells, butI am far from being convinced that this is a right view of thecase.1

If there is one thing more than another which has comeinto prominence as the result of recent research, both botani-cal and zoological, it is the fundamental importance of thenucleus to cell life. So many minute organisms, which at one

1 With regard to the argument which follows, I would remind my readersthat Hackel, thirty years ago, clearly expressed the view which I am nowurging (see his " Allgemeine Anatomie den Organismen," forming the firstpart of the ' Generelle Morphologie,' p. 296). " Es muss hierbei ausdriicklicherrinert werden, dass wir unter eine Zelle nur einen Plasma-Klumpen miteinem Kerne verstehen konnen. Der haufig gebrauchte Ausdruck einer' mehrkernigen Zelle' ist eine Contradictio in adjecto, da ja eben nur dieEinheit des Kerns die individuelle Einheit der Zelle als eines Elementar-Organismus bedingt. Jeder Plasmaklumpen, der mehr als einen Kernumschliessfc, moge er nun von einer Membran umhiillt sein oder nicht, ist eineVielheit von Zellen, und wenn diese Yielheit eine bestimmte einheitlicheForm besitzt, so haben wir sie als Zellenstock zu dem Range eines Organeserster Ordnung zu erheben." This view, however, has been controverted bymany authorities, as will appear further on.


time were believed to be non-nucleate, have since been shownto contain nuclei, or at any rate nuclear matter, that we aretolerably well justified in saying that the nucleus, or itsequivalent, is an essential constituent of the cell. At allevents we know that division of the nuclear substance, whe-ther mitotic or amitotic, is all-important as a prelude to andaccompaniment of cell division. The experiments of Gruberand Verworn show that if Amoebae are artificially divided, theparts cut off will regenerate and lead an independent existenceif they contain nuclear matter, but if they do not, they soonperish. Fragmentation of the nucleus—by which is produceda so-called multinucleate condition, often of considerableduration—is a prelude to spore formation, i. e. to thedivision of the cell into many parts. Mitotic division is highlycharacteristic of division of the cell into two parts. It is verydifficult to draw distinctions, but it is worth considerationwhether the temporary multinucleate condition ending inmultiple fission, which is common in protozoa, has not adifferent value to the permanently multinucleate condition ofsome plants and animals, which are generally called unicellular.In the one case(e. g.Podophrya,Thalassicolla,Actinosph8erium)division or fragmentation of the nucleus leads, sooner or later,to the separation of cells, each containing a fragment of theoriginal nucleus. In the Coeloblastse (Siphonese, e. g. Caulerpa)the repeated division of the nucleus is not followed by anycell division, but the organism is throughout life a mass ofcontinuous undivided protoplasm. The plant, as von Sachssays, is of considerable size, develops roots, even leaf-formingshoots, and in its protoplasm hundreds and thousands of cellnuclei are contained, which with advancing growth are multi-plied by division, and obtain a definite arrangement within theprotoplasm. And, as in the case in cellular plants, the nucleiare specially aggregated at the growing points. The wholebehaviour is just that of a multicellular plant, but there are nopartition walls.

It is stretching the point very far to call this a single cell.And, in fact, it is an inconsistency to do so, for where, by an


essentially similar process, a continuous sheet of protoplasmcontaining many nuclei is formed as a tissue-constituent of amulticellular animal or plant, we do not call the whole multi-nuclear tract a single cell—we call it a syncytium, or takesome roundabout way of describing it. Such a case is theformation of the endosperm in the embryo-sac of Phanero-gams. By repeated mitotic division of the nucleus and growthof the surrounding cytoplasm, a tract of continuous proto-plasm is formed, containing many nuclei. At a later stagepartitions are formed and the mass is divided up into cells, butfor a period the endosperm has a structure which recalls thatof the Coeloblastse. Can we say that the condition in theendosperm is to be regarded as multicellular because it is notpermanent, and that the condition in the Coeloblastse is to beregarded as unicellular because it is permanent ? If this isallowed the consequences are far-reaching, for it follows thatthe multinuclear phase in Actinosphserium and other Protozoais also multicellular, because not permanent.

Take, again, the case of the Mycetozoa. The plasmodium ofBadharaia or Fuligo is not unicellular, for it is formed by theunion of many cells: it is not called multicellular, becausethere are no cell divisions : yet we draw, rightly enough, adistinction between the plasmodium, where cell bodies fuse butthe nuclei do not unite, and the single cell resulting from, con-jugation, where the nuclei do unite.

A survey of the facts must lead to the conclusion that thereis an intermediate phase between the unicellular and the multi-cellular condition, which is the multinucleate but non-cellularcondition,1 and that there is no fundamental distinction

1 The term non-cellular does not exactly represent the condition which itis intended to describe. Yet, if one adheres to existing nomenclature, it isdifficult to find a substitute. The term "cell," though founded on anerroneous conception, is so firmly established in biological language that itwould probably be impossible to eject it. Yet if one were to make generaluse of the Greek equivalent KVTIS (literally a little box), which has alreadycome into such favour as to have respectable claims on our attention, onsmight adopt much more exact expressions. Thus the uninucleate Protozoamight be said to exhibit a monocjtial condition, multicellular organisms a


between Protozoa as unicellular, and Metazoa as multicellularorganisms. I should hardly have thought it worth while toinsist upon this had not Mr. Sedgwick written "that anorganism may consist of one cell or of several cells in associa-tion with one another. We draw the most fundamental dis-tinction between the two kinds of organism, and we divide theanimal kingdom into two great groups to receive them. As aproof of the importance which we attach to this feature oforganisation we assert that a man is nearer, morphologically,to a tapeworm than a tapeworm is to a paramcecium."

Botanists, who have the great advantage of studying thephysiology concurrently with the morphology of their subject,make no fundamental division into Protophyta and Metaphyta.For them, unicellular plants, hypopolycytial plants, Fungi andAlgae are alike Thallophyta, and a passage from Goebel mayserve to illustrate the point of view which leads them to classifytogether organisms which, from the point of view of "inde-pendent life units," would appear widely separate. "Fromthis initial stage "—a single small cell—" the process of de-velopment may advance, yet still within the limits of a singlecell, and whilst the cell increases in size, often reachingdimensions without parallel in the vegetable kingdom, eitherthe differentiation of the cell-contents or that of the externalform, as shown by the branching, may make most rapid pro-gress. In other cases the .growth of the cells is accompaniedby cell-division, the thallus becoming multicellular, and thesingle cell producing, according to the nature of the plant,a cell row, or a cellular filament, a cell surface or simple tissuelayer, or lastly a cell mass increasing in every direction.'"

polycy tial condition, and tbe so-called non-cellular condition of Cceloblastseand Opalina might appropriately be called hypopolycytial, the prepositionv7ro being used in a modifying sense, as expressing the intermediate stagebetween one and many. The term syncytial, which is now used in a loosesense, is strictly applicable to the early condition of the plasmodia of theMyxomycetes, which are formed by the fusion of many units in a monocytialcondition, and are therefore different from organisms which exhibit a hypo-polycytial condition. In later stages the nuclei of the plasmodia multiply bydivision; thus the hypopolycytial is added to the syncytial condition.

A CRITICISM OP THE CELL-THEORY. 155

Although in this passage, which is descriptive of Thallophytes,Goebel attaches too much importance, as I think, to the con-tinuity of a vesicle as determining the unicellularity of a plant,he shows clearly enough that he regards the growth and modeof extension of the protoplasm, not its division into cells, asthe feature of fundamental importance.

There is the further property in plants that continuitybetween the cells of highly organised multicellular plants hasbeen shown to be of very general, if not universal, occurrence.And if complete separation were to be insisted upon as acharacteristic of a cell, any given Angiosperm, or other highlyorganised plant, could no longer be considered as an aggregateof life units, but rather as a conjunct mass of protoplasm,imperfectly broken up into corpuscles, in each of which thereis a nucleus. It is but a step from the much-branched, multi-nucleate Coeloblastse, which have no partitions, to the forma,tion of incomplete partitions, breaking up the protoplasm intosmall masses, which remain, however, linked with one another,and so preserve an original continuity similar to that of theCoeloblastse, which has only apparently but never actually beenbroken.

So much has this idea impressed itself on the minds of someobservers, that Hofmeister suggested that the creeping motionof the plasmodia of the Myxomycetes and their later transfer,mation into fructification, is representative of the simplest typeof growth, even for more highly organised plants. Thisopinion has been quoted with approval by von Sachs, who,before even the continuity of the protoplasm of plant cellswas established, wrote that " fundamentally every plant, how-*ever highly organised, is a protoplasmic body, coherent initself, which, clothed without by a cell-wall and traversedinternally by innumerable partitions, grows; and it appearsthat the more vigorously this formation of chambers andwalls proceeds with the nutrition of the protoplasm, the higheralso is the development attained by the total organisation."

Expressed in this way, the phenomenon of cell-formation isrepresented to us as being nothing more than a particular


manifestation of growth, and Mr. Sedgwick may contend thathis views are thereby conceded, and that the ancestral meta-zoon may, on this aspect, be considered as " a multinucleateinfusorian with a month leading into a central vacuolated massof protoplasm." There may be truth in the contention, yetnone the less we may hold fast to the concept of a cell, as I shallattempt to show further on. And it may be observed in pass-ing that Mr. Walter Gardiner, in describing and emphasisingthe continuity of protoplasm in plants, expressly stated " thatthe presence of minute perforations of the cell-wall need notlead to any modification of our general ideas as to the mechan-ism of the cell," a proposition which most reflective personswill be cordially inclined to agree with. For this much iscertain, that the formation of cells is not merely the expressionof one out of many formative processes which reside in organicmatter, but is the formative process, par excellence, whichobtains both in animal and vegetable tissues.

Thus far I have endeavoured to show that the independent-life-unit theory has not held the minds of zoologists in an ironbondage, much less the minds of biologists, for, when referenceis made to biologists, botanists must be taken into equalaccount with zoologists.

It is, however, arguable that, whatever botanists have thought,zoologists have not followed their example, but have publiclymaintained a complete adherence to the independent-life-unittheory in its most limited form, whatever reservations theymay privately have made in their own minds.

But it may be doubted whether the argument holds good.1 have already shown that passages which seem to state mostdogmatically that cells are separate individuals, prove onexamination to be nothing more than illustrations ; and it isto be.remembered that ideas founded on botanical evidencemust always be reflected on the minds of zoologists, and onemay certainly say that conceptions of animal structure haveof late years been considerably modified by the light thrownupon organic structure in general by botanical investigation.Some zoologists may possibly have given too little attention, to


growth without division into cells, because there are not inthe animal kingdom any such striking instances of massivegrowth without cell division as are exhibited by the Coeloblastse,especially if we leave out of consideration the Mycetozoa, asbelonging to the debateable territory between the two kingdoms.Nevertheless, we have instances of growth and mitotic nucleardivision, unaccompanied by cell division, which are not ap-parently a mere prelude to division. Take the single instanceof Opalina r ana rum. Because this organism is microscopic,and may be described, without offence to our sense of propor-tion, as a corpuscle, it is invariably called unicellular. Yet inessential features it resembles one of the Coeloblastee. Itcontains numerous nuclei, which divide mitotically, and theirdivision is an accompaniment of the growth of the matureorganism. Themultinucleate mature condition is of considerableduration. In the reproductive process this multinucleatecorpuscle divides repeatedly, until a number of small offspringare formed, each containing several, usually four or five, nuclei.The minute product of fission then encysts, and it is remarkablethat either during or immediately after encystment the severalnuclei break up, and a single new nucleus is formed,—pre-sumably it is constituted out of the chromatin of the severalnuclei. The form which emerges from the cyst grows, andgrowth is accompanied by repeated mitotic division of thenucleus till the mature condition is reached. The wholehistory reminds one of that of a Mycetozoon, except that theyoung do not fuse to form a plasmodium, but simply grow up;in this respect Opalina resembles the Coeloblastae, differing fromthem, however, in the fact that the whole organism is concernedin reproduction, not a special part. Although it has, as heremarks, a distinct "development," Zeller, who first followedits life history, has no doubt that Opalina is a single cell.

Now the multinucleate condition is far from uncommon inthe Protozoa, and it may almost be said to be the rule in theCiliata, if we regard macrouucleus and micronucleus as twoseparate nuclei. But putting aside this phenomenon, thesignificance of which we do not yet clearly understand, there


are several Ciliata which have as many as one hundred nuclei,e.g. Holophrya oblonga, Lagynus e longatus , andUroleptus roacovianus.1 I do not include as multinuclearthoseformsinwhich,asiuTrachelocerca phoeni copter us orChcenia teres, the chromatin is scattered throughouttheproto-plasm in the form of minute granules. Those Protozoa only maybe considered multinucleate in which there are several well-defined aggregations of chromatin. And even if the Ciliataabove mentioned may not be considered truly multinucleate, butto possess onlya fragmented nucleus,there can be no doubtaboutsome Amoebae, e.g. Amoeba quarta and others described byGruber.2 In the last-named the multinuclear state is con-stant ; as Gruber says, " es sich nicht etwa urn vorubergehendeEntwicklungszustande handelt." He watched these Amoebaefor a long period, expecting that the large number of nucleiwould at last find its explanation in reproduction by multiplefission, but he was unable to observe any such culmination.Dr. Gruber is a great authority, and he, equally with Zellerand others, is quite positive that the multinuclear Protozoaare truly unicellular. His reasons are, that closely alliedspecies are uninuclear, and that the protoplasmic body is con-tinuous—contained in the case of Ciliata by a single cuticularcoat. But even he admits that the only reasonable interpreta-tion of the multinuclear condition is that it is a prelude toreproduction, that is, to cell division.3 It is, therefore, a con-dition intermediate between the unicellular and the multi-cellular condition, or, as I should like to call it, a hypopoly-cytial condition, and nothing more need be affirmed of it.

Zeller is quite precise as to his reasons for regardingOpalina as unicellular. " Die kleinsten. Thierchen aller be-kannten Opalinen, so wie sie von Neuem sich zu entwickelnbeginnen, besitzen nur einen einfachen Kern und entsprechen

1 Maupas, "Etudes des Infusoires cilies," 'Arch. Zool. exper. et gen.' (2),i, 1883.

2 ' Zeit. fur wiss. Zool.,' xli, p. 186.' Aug. Gruber, "Ueber vielkernige Protozoa," 'Biol. Centralblatt,' iv,

p. 710.


unzweifelhaft, wie Engelmann schon fiir die von ihm untersuchteArt nachgeweisen hat , ' morphologischvollstandigeinereinzigenZelle.' Aber auch mit der weiteren Entwicklung andert sichdaran nichts. Mag die Zellhaut zu einer aus vielen einzelnzerlegbaren Bandern bestehenden muskuloses Hiille werdenund mag der Kern in zwei Kerne zerfallen, wie in O. similisund O. caudata , oder durch fortgestzte Theilungen eineschliesslich sehr grosse Menge von Kernen aus sich hervorgehenlassen, wie in O. ranarum, 0. obt r igona und 0. dimidiata ,die protoplasmische Korpersubstanz selbst zeigtke ine weitere Veranderung als die der Masseu-zunahme und blebt, wie auch Engelmann hervorhebt ,' Ze i t lebens eine einzige zusammenhangender Masse,•wie von eine einzigen Zelle.5" I have put the last passagein italics, because it expresses most clearly why Zeller and otherauthors regard multinucleate forms as unicellular, namelybecause the protoplasm showa no other change than increasein size, and because it remains, its life long, a single con-tinuous mass. The same argument leads many to regard theCoeloblastae as unicellular. The continuity of the protoplasm,then, is the test of unicellularity.

If anybody accepts this, he cannot escape from its logicalconsequences. Not only are multinucleate Protozoa andCoeloblastae unicellular, but also the whole kingdom of plants,for their protoplasm is continuous: the developing Peripatusis unicellular, for its protoplasm is continuous ; the epithelialcells of many animals, as Max Schulze, Pfitzner, Klein,Paulicki, Th. Cohn, and others have shown, are united by fineprotoplasmic processes much as are the cells of plants, thereforethe epithelia are unicellular, for their protoplasm is continu-ous. The same may be said for muscle cells (Werner andKlecki), for connective tissue, for bone cells, for the developingmesoblast of Vertebrata ( teste Sedgwick, Assheton, andothers), for the mesoblast (mesenchyme) of trochospheres andMolluscan larvae (see particularly von Erlanger), and for manyother tissues.

Thus the inevitable result of an argument which is meant


by those who use it to tighten the bonds of the cell-theory is.to loosen them altogether, and to hand us over unbound toMr. Sedgwick, who would fetter us once more with a newdoctrine, viz. there is no cell, all organisation is a specialisationof tracts and vacuoles in a continuous mass of vacuolated pro-toplasm.

We do not want to be bound, at least I do not, and if we areto be free we must take refuge in some such lax but compre-hensive statement as that of von Sachs, viz. that cell forma-tion is a phenomenon very general in organic life ; but even ifwe must regard it as only of secondary significance, it is thecharacteristic expression of the formative forces which residein organic substance.

Now this statement affirms the existence of cells, and it isnecessary to arrive at some understanding as to what is a cell;what properties are connoted by this term ?

It has become abundantly evident in the course of thisargument, that whatever other attributes may be affirmed ofthe cell, the possession of a nucleus is one of the most im-portant. It is impossible to disagree with Pfitzner when hewrites, "Wenn wir aber den Kern iiberall und zwar immerund in alien Stadien als durchaus selbstiindiges Gebilde finden,so ergiebt sich deraus dass er fur das Bestehen der Zelleals solchen ein Organ von wiel fundamentaler Bedeutung istals wir bisher geneigt werden anzunehmen." This is also theview of 0 . Hertwig, and it is no new one, for Max Schulzeinsisted upon it, and Hackel wrote in 1866, "Ein Plasmak-lumpen ohne Kern ist keine Zelle mehr."

But can we follow Pfitzner when he goes further and says," bei einer so ausserordentlich Konstanz in der ganze Reiheder Thierformeu, von den Protozoen bis zu dem Menschen,kann ich nicht umhin auzunehmen dass iiberhaupt die ganzeExistenz eine Zelle als biologische Einheit an das Vorhandseineines centralen Korpers, von komplicirten inneren Bau,gebunden ist, dass also die Chromatinstrukturen nicht etwassekundaren erworbenes, sondern die Grrundbediugung vitalerExistenz der Zelle darstellen. Uud weiter folge ich hieraus


das der als Karyokinese bezeichnete Vorgang nicht ein spe-cielle Kerntheilungsmodus, sondern der Kerntheilungsmodustear' E^o^/jv ist " ?

I think not. Particles of chromatin scattered through theprotoplasm do not constitute a nucleus any more than a heapof bricks constitutes a house. Under such a view, Ciliata likeTrachelocerca phsenicopterus and Chaenia teres wouldnot be cells, for they have no central nucleus of complex struc-ture, nor have Oscillaria and Bacterium, in which chromatingranules have been discovered. Though the case of Holos-t icha scutellum, in which scattered nuclei (chromatin par-ticles) unite and fuse to form a single central body or nucleusprevious to division, may help to clear our ideas, it is evidentthat the demand for a cent ra l organised cons t i tuent ismore than the cell conception can bear, especially if thedemand carries with it a further demand for the universalityof mitotic division in nuclei.

In short, before we could accept Hertwig's definition ofa cell, we should have to ask and answer the question, Whatis a nucleus?

Here I may stop to ask whether it is worth while to discussthe grounds of a definition which, when made, could not beacceptable to the mind of everyone. An argument aboutdefinitions would soon land one in the regions of scholasticism,and I have no desire to enter into subtleties which would taxthe powers of a Duns Scotus. To give an answer which shallbe beyond all cavil to the question, What is a nucleus ? wouldbe about as easy as to answer how many angels can dance onthe point of a needle.

The truth is that it is the attempt to frame short concisedefinitions, applicable without exception to whole classes ofphenomena, which leads to trouble. The concepts of biologymay and should correspond with the phenomena we observe,but they can very seldom be made into universal propositions.There is no place in the science for definitions as exact anduniversal as those of geometry. The qualities of a nucleus arenot to be defined like those of a point or a line. Such propo-

VOL. 3 8 , PART 1. NEW SERIES. L


sitions as we may make are but resting-places for our mindsas we ascend the mazy scale of organisation. To attempt toform definitions, to predicate the precise attributes of wholeclasses of phenomena, is to run counter to the very genius ofthe subject. For what do we mean by evolution if not thatlife is labile, never resting, protean in its variety? And howcan we express this but in an incomplete way, contenting our-selves with particulars, and trying to show that the stream,though it flows in many tortuous channels, is one streamnevertheless.

Cells and nuclei are protean in their variety, and since wevery rightly insist on objective study as a preliminary to theunderstanding of them, it is not wonderful that they shouldgive rise to this concept in the mind of one man, and to thatconcept in the mind of another man, and thus it is not sur-prising that the theory of cells should be incapable of beingstated, as Mr. Sedgwick complains, " i n so many words in amanner satisfactory to everyone."

It is fairly obvious that Mr. Sedgwick's quarrel with thecell-theory began with the dissatisfaction which he felt whenhe discovered that doctrines, which he believed to be of uni-versal application, were in fact contradicted by several instances.But he fell out of Scylla into Charybdis when he supposedthat he could reply to a uuiversal affirmative by a universalnegative.

There is an old and respectable rule of logic that of twocontrary propositions both cannot be true and both may befalse, whilst of two subcontrary propositions both may betrue but both cannot be false. Had Mr. Sedgwick remem-bered this, he would not have attempted to overthrow thecell-theory by the statement of a contrary proposition ofequally universal import.

The cellular theory of development in the popular form inwhich it is often presented may be briefly summed up some-what as follows. The multicellular organism is a colony, con-sisting of an aggregation of separate elementary parts, viz.cells. The cells are independent life units, and the organism


subsists in its parts and in the harmonious interaction of thoseparts.

The falsity of this summary is evident when we considerthe known facts of vegetable organisation; the developmentof Peripatus; the union, by means of protoplasmic processes,of epithelial, muscular, and connective-tissue cells; the evidencelately adduced as to the continuity of the mesoblast in Elasmo-branchs, Aves and Mammalia, and other well-known instances.

The absolute contrary, as expressed by Mr. Sedgwick, isequally false, viz. that the metazoon is a continuous mass ofnucleated vascular protoplasm, subsisting in the unity of itsmass. For, as I have shown in the earlier part of this essay,there are unequivocal instances of distinct isolated cellsoccurring in the embryos of many Metazoa (Nereis, Unio,Umbrella, Leptoplana). Moreover I am convinced, by my ownstudies on the histology of Coelenterates, that, whilst there isorganic connection between many of the tissue-cells composingthese organisms, as was demonstrated long ago by the brothersHertwig, there are many other cells of which such continuitycannot be affirmed.

To deal clearly with the cell-theory, or rather with the inde-pendent-life-unit theory which has grown out of it, we mustsplit it up into as many separate propositions as it contains.These are:

The multicellular organism is an aggregate of elementaryparts, viz. cells.

The elementary parts are independent life units.The harmonious interaction of the independent life units

constitutes the organism.Therefore the multicellular organism is a colony (cell-

republic according to Hiickel).It is not necessary to follow the theory further into the

consequences which are deducible from these propositions,e. g. that development consists in the separation of numerousindividual units from a single primary unit, the ovum. It isobvious that the truth of the first proposition in no waydepends on the truth of those which follow, and that, in fact,


the second proposition is an assumption which is made toexplain the first. We may make Mr. Sedgwick a present ofthe last three, whilst we retain and value the first.

The essence of the whole question is this: are we justifiedin considering the elementary parts of an organism to beindependent life units ? Before we can answer this, we mustinquire why we do consider them to be independent life units ?

The answer to this is probably to be found in the aphorism,which commends itself to everybody, that reproduct ion isdiscont inuous growth. From the observation that, inunicellular organisms, division of the unit—the cell-corpuscle—leads to the liberation of a new and independent unit, andthat in multicellular organisms it is the liberation of anindependent unit—the ovum—which constitutes reproduction,it has become a settled conviction in men's minds, that divisionof a cell-corpuscle means the liberation of a new unit, that is}

the setting free of a new independent being. It is this con-viction which has led to the belief that the units composing amulticellular organism are in posse independent beings,though in esse subordinate to the whole of which they forma part. This was the argument of Schwann when he wrotethe passage which I have quoted on p. 149, and the argumenthas been taken as conclusive.

But we know now that the power which Schwann and hisfollowers limited to cells is inherent in protoplasmic massesnot divided into cells. For instance, if the cell-membraneof a Coeloblastic alga is ruptured, portions of the exudedprotoplasm, provided they contain one or more nuclei, maybecome, after a time, surrounded by a new cell-membrane,grow, and form a new plant.

The experiments of Gruber show also, that portions ofAmoebae artifically separated may, provided that they containnuclear substance, recover from the operation, and lead anindependent existence;

May I ask, in parenthesis, whether there can be a betterillustration of the truth of the contention which I have en-deavoured to establish above, that whilst a uninucleate cor-


puscle of protoplasm is in esse as also in posse a unit ofindependent vitality, a multinucleate corpuscle or mass ofprotoplasm is in posse composed of separate cells (units ofindependent vitality if one chooses to call them so) whilst stillin esse a single unit of independent vitality ?

To continue the subject. We now know also that divisioninto cells is not necessarily, though it sometimes may be,division into units of independent vitality, but is often (may wenot say generally ?) incomplete separation into form elementswhich may indeed, under certain conditions, be completelyseparated, and exhibit an independent vitality (Begonia), butunder normal conditions participate in the vitality of the wholeplant or animal by means of their connections with theirfellows. Hence we must conclude, as it seems to me, that theelementary parts of organisms are not independent life unitsin esse. They may be so in posse in many cases, but asdifferentiation and specialization progress they lose this poweralso, and cannot, when separated from the whole of which theyform a part, exhibit independent activities.

This consideration leads to the apparent paradox, that thehigher the organisation the less conjunct and, at the sametime, the less independent are its parts ; the lower the organisa-tion the more conjunct, but also the more independent are itsparts.

This is a puzzle which has, for years past, exercised theminds of biologists. There is, I believe, but one solution ofthe difficulty, and it is to be found in the physiological importof cells.

But before we can enter into this question we must finallysatisfy ourselves, as far as circumstances allow, about themorphological concept of a cell.

That the cell is a thing cognisable, and that it is not anunreal figment, due to imperfect observation or to hopelesslyprejudiced interpretation of our observations, as Mr. Sedgwickwould make us believe, I will try to show. •;

A cell is a " body," and therefore an external cause to whichwe attribute our sensatipns. I would submit that, without

166 GILBERT C. BOURNE.

prejudice to the metaphysical standpoint, we must conceivethat what is capable of giving rise in us to such very distinctsensations, must have a real existence. I am referring now tothe component'parts of the tissues of higher animals and plants,and not to unicellular organisms.

If, then, the thing has existence, it must have attributes;we must be able to affirm something of it. What we have toaffirm is not the attributes of this cell or of that cell, but ofcells in general. We have to give expression to a morpho-logical idea, in the sense in which Goethe used the wordmorphological. Our concept of a cell must be an " Allgemeinesbild," the generalised idea of a cell, derived from our ex-perience of many kinds of cells. I have already shown, atsufficient length, that we must now regard something of thenature of a nucleus as an essential component of all cells, butas the concept of a nucleus as a central organised body is notapplicable to all cells, I would widen Max Schulze's definitionby saying that " a cell is a corpuscle of protoplasm, whichcontains a specialised element, nuclein." This is a sufficientlycomprehensive statement of our " Allgemeines bild," thoughI cannot pretend that it is not open to objection.

Cells, as thus defined, are not only of various kinds, butthey are variously compounded together. We may, by theprocess of dichotomous division, classify them, according totheir relations to other cells, as discrete and concrescent.

By discrete cells, I mean those whose protoplasm is notin union with that of any other corpusde.

By concrescent cells, I mean corpuscles whose protoplasmis in union with that of other corpuscles.

Discrete cells may further be divided into:Independent cells, living wholly apart from one another,

or separated by an appreciable interval of space, e. g. uni-nucleate Protozoa, the mature ovum, leucocytes.

Coherent cells, which are in close apposition to others,but not organically in union with them, e. g. the blastomeres ofmany developing embryos.

Concrescent cells may also be further divided into :

A CRITICISM OF THE OELL-THEORT. 167

Cont inuous cells, whose protoplasm is fused but whosenuclei are separate, e.g. Myxomycetes, Cceloblastae,Opaliua.

Conjunct cells, those which having a protoplasmic bodyof definite outline are united i n t e r se by fine bonds of proto-plasm, e. g. vegetable tissue cells, epithelial cells of manyanimals; meseuchyme cells, &c.

Experience shows us that independent cells may, in processof growth, give rise to coherent cells, continuous cells, con-junct cells, or to all three together, and that coherent, con-tinuous, or conjunct cells may, and in fact do, give rise toindependent cells. As thus stated, can there be a betterillustration of von Sachs's principle that cell-formation is anaccompaniment of growth ?

It will be observed that, in adhering to the present termi-nology, I am obliged to classify organisms usually (though notalways) called unicellular as multicellular. I have tried toescape from this necessity, but the limitations of languagecompel me to it. I should be grateful for a better and morelogical definition.

The view of Mr. Sedgwick—if I do not misrepresent him—is this, that there are no coherent cells; that all which I haveclassified as continuous and conjunct cells are not cells, buttracts of protoplasm; that the only cell, sensu str icto, isthe independent cell, and that morphologically and physio-logically it is of no consequence.

I have already shown that there are cells which we mustregard as coherent. I cannot, for reasons which I will explaindirectly, consider the independent cell of no consequence, andthe difference between us as to conjunct cells is simply this ;Are they to be regarded as one or many ? I can, perhaps, bestexpress this difference by an illustration.

Is a house to be regarded as one room or composed ofseparate rooms ? A room is a certain portion of space enclosedby walls, ceiling, and floor; but it is also in connection, bymeans of the door, with other similar rooms. Is it, then, nota separate room, but part of a larger room ? Or if I shut the

168 GILBERT 0. BOTJENE.

door is it a room, and if I open the door is it no longer aroom ? The subject might be argued with much ingenuity,but the final answer is this—that " room " and " cell " areterms which give expressions to certain states of our conscious-ness, and for practical purposes they are very useful termsindeed. Where distinct states of consciousness are called up,of such a nature as to give rise to ideas of par t icular i ty , itis a mere quibble to argue that the apparent parts are actuallymerged in a whole. A cell is none the less a cell, in the senseof a thing distinct in itself, because it is conjunct with itsfellow cell, than my room is the less a room because it hasone door opening into an adjoining room and another openinginto the passage.

Yet there is something more than a verbal quibble in Mr.Sedgwick's contention. He would have it that in the case ofmesenchyme it is incorrect to say that it is a number of stel-late cells joined to one another by their processes. For himthe correct description is, " a protoplasmic reticulum withnuclei at the nodes." Does he accept the logical consequencesof this, and say of the epithelial cells of the Salamander or ofunstriped muscle fibres that they are protoplasmic reticulawith nuclei at their nodes? And if so, how does he explainthe fact that, in the one case and in the other, the elementswhen absolutely isolated by appropriate methods show a re-markably constant and characteristic form ? Were they whathe describes, rupture of the internodes of the reticulum wouldresult in amorphous lumps of protoplasm, not in units ofcharacteristic form. It is the constancy of the various formsof cells which convinces morphologists of their individualityas form elements, and all the arguments which Mr. Sedgwickor anybody else may choose to bring forward will not convincethe man who goes into a laboratory, makes a few macerationpreparations, and studies the results for himself.

Thus a tissue formed of conjunct cells is made up of manyand not of one, and as a form concept the cell holds its groundand, pace Mr. Sedgwick, it will continue to hold its groundagainst all comers.


As a physiological concept it is hardly less useful, thoughreflection may induce us to abandon the " cell-republic"theory, as, indeed, it has been tacitly abandoned by many.I take it that the scheme of von Sachs very nearly expresses,in general terms, the physiological importance of the cell.An organism is a protoplasmic body, coherent in itself, whichgrows, and as it grows it is divided by cleavage into innu-merable corpuscles, and it appears that the more vigorouslythis formatian of corpuscles proceeds with the nutrition of theorganism, the higher also is the development attained by thetotal organisation. Nor does this statement stand in any con-tradiction to the original theory of Schwann, from whom Imay quote two more passages : "The elementary parts of alltissues are formed of cells, in an analogous though verydiversified manner, so that it may be asserted that there isone universal principle of development for the elementaryparts of organisms, however different, and that this principleis the formation of cells." And again, he says of the relationsof cells to one another, " Each cell is within certain limits anindividual, an independent whole. The vital phenomena ofone are repeated, entirely or in part, in all the rest. Theseindividuals, however, are not ranged side by side as a mereaggregate, but so. operate together in a manner unknown tous, as to produce a harmonious whole." It should be remem-bered that Schwann regarded cells as so many separate vesi-cles, and when allowance is made for this error, the secondpart of the last passage must be allowed to have great signifi-cance. The subordination of the parts to the harmoniouswhole, leading to the loss of individuality of the parts, inanimal tissues, was insisted on by Hackel in his ' GenerelleMorphologic' The first of the two sentences which I havequoted from Schwann is even more true to-day than when itwas written, for we have got rid of the cell-forming matrix,the cytoblastema ; and I would wish to insist on this passageas expressing in the clearest possible language the cell-theoryas we understand it to-day.

From this standpoint we can see, obscurely it may be, why


cell-formation accompanies differentiation with growth of themass, and why specialisation is not possible in continuoustracts of protoplasm. For, as Mr. Sedgwick himself admits,in a continuous mass of protoplasm, changes of molecularconstitution in any one part would in time spread throughthe whole, so that a differentiation of one part would in timebe impressed on all the other parts, and physiological divisionof labour would be out of the question. The fact that in theProtozoa there is differentiation within the limits of a singlecorpuscle presents no greater difficulty than the fact that inthe epithelio-muscular cells of Coelenterates, or the similarcells in Nematodes, there is differentiation within the limits ofthe cell.

Again, metabolism in a large mass is greatly facilitated byits being broken up. As von Sachs says, " It is very intelligiblethat not only the solidity but also the shutting off of variousproducts of metabolism, the conduction of the sap from placeto place, and so forth, must attain greater perfection if thewhole substance of a plant is divided up by numerous transverseand longitudinal partitions into cell chambers." The samething applies, muta t i s mutandis , to animals, and it is notdifficult to see that the difference between holozoic and holo-phytic nutrition makes it impossible for. the animal to growto a large mass without division into cells, whilst such growthis possible in the case of plants which, like Codium andCaulerpa, live in water, or like Botrydium in damp earth.

It is known that the spaces between epithelial cells whichare traversed by the connecting strands of protoplasm, andwere formerly supposed to be occupied by a cement substance,are in reality lymph spaces, and this gives us some insight intothe importance of the cell structure in animal organisation.The formation of cells with spaces between admits of nutrientfluid being brought to the very threshold of each constituentcorpuscle of the organism. (See on this subject Th. Cohn,R. Heidenhain, Paulicki, Nicolas, Werner, and others.)

Whilst the necessities of cohesion, solidity, and transmissionof stimuli may explain the conjunct nature of so many tissue

A ORTTIOISM OP THE OELL-THEORY. 171

cells, recent researches on cell lineages may perhaps give usa clue to the interpretation of the fact that blastomeres arein so many cases, no more than coherent. For it is noticeablethat wherever cell lineages, with marked isolation of the blasto-meres, have been described, there is a decided tendency to theprecocious development of organs, or, at any rate, to the pre-cocious isolation of the primordia (Anlage) of organs.

It seems probable that the discrete condition of the blasto-meres is connected with the fact, to which I alluded in theearlier part of this essay, that they are, from the very outset,specialised. They have each a definite molecular constitutiondifferent from the others, and, in figurative language, alimited part to perform, which they could not perform toadvantage if they were conjunct with the other blastomeresand shared in their different molecular constitution. Butthis is a subject which I must leave for a future occasion whenI discuss the validity of von Baer's law of development.

( I have travelled in this essay over a great deal of ground,and I have necessarily had to touch more lightly on manytopics than I should have wished. I hope that I may atleast have succeeded in presenting my arguments in a mannerwhich will make them clear to my readers, and that I have notbeen too discursive. Starting from Mr. Sedgwick's propositionsand accusations, I have tried to show what is or was the exactextent and meaning of the cell-theory ; I have tried to examineit and show how much was good and how much bad, and Ihave finally been led to the conclusion—which is not quitewhat I proposed to myself at the outset—that the cell conceptis a valuable expression of our experience of organic life, bothmorphologically and physiologically, but that in higher or-ganisms cells are much what von Sachs declares them tobe, not independent life units (Lebenseinzelheiten), but aphenomenon so general as to be of the highest significance;they are the constant and definite expression of the formativeforces which reside in so high a degree in organic matter.

Lest I should appear to have minimised the importance ofthe cell too much, let me conclude by saying, that nothing


which has appeared above calls into question that great featureof animal and plant development which most impresses thebiological student, viz. that organic growth is a cycle, beginningin the single cell, and returning to the single cell again. Andtherefore, in a limited sense, the cell is par excellence theunit of life. Its growth takes various forms and shows manycomplexities, but whatever the form, however great the com-plexity, it is a progress from the state of an independentcorpuscle, through a state of many coherent, or continuous,or conjunct, interdependent corpuscles, back again to the stateof a single independent corpuscle.

This was the great advance made by Remak on the theoryof Schwann, and summed up in Virchow's aphorism, which Ibelieve to be universally true. For Schwann did not hold thatcells are the ultimate basis of life: he held that they areformed, as a crystal is formed out of its mother liquor, from astructureless matrix, the cytoblastema. To some such theoryMr. Sedgwick wishes to take us back again, for his " pale and-at first sparse reticulum " bears a most suspicious resemblanceto the exploded cytoblastema. " The development of nerves,"he says, " is not an outgrowth from certain central cells, butis a differentiation of a substance which was already inposition." And earlier in his article, referring to the growthand extension of the mesoblast between epiblast and hypoblast,he says : " What are the facts ? The space between the layersis never empty. It is always traversed by strands of a paletissue connecting the various layers, and the growth whichdoes take place between the layers is not a formation of cellsbut of nuclei, which move away from their place of originand take up their position in this pale and at first sparsereticulum."

But surely nobody ever affirmed that the space between thelayers was empty except in the sense that it is devoid ofcellular structures. It is well known that it is filled with acoagulable fluid, and it is worthy of remark that coagulablefluids, treated with the reagents now most in use, frequentlyform a reticulum of pale non-staining substance. I can speak


from experience, for not long since I was much puzzled bysuch a reticulum, and had I been less cautious I should havepublished, as a great morphological discovery, statements whichrested on a wholly insufficient basis of experience. The subjectrequires further investigation, and the most that one can saynow is, that it is possible that Mr. Sedgwick, good observer ashe is, may have been mistaken. And he will pardon myobserving that the things which he states are not " facts."They are his own inferences from his own individual observa-tions, and will require very abundant confirmation before theycan take rank as what we agree to regard as " facts." All the" facts " we have at present, i. e. the accumulated observationsof hundreds of highly-trained and able observers, are funda-mentally opposed to any such account of protoplasmic growthapart from nuclear formation as Mr. Sedgwick gives us. Butthere is another way of looking at it, namely, that he has onlyoverstated his case, and that the growth of the tissues inquestion resembles the apparent creeping motion of the plas-modia of the Myxomycetes. That this may be the case issupported by a study of Mr. Assheton's recent account of thegrowth of the mesoblast and of the inner layer of the epiblastin the embryo of the rabbit. It presents no theoretical diffi-culties, but it should be remarked that Mr. Assheton figuresnumerous nuclei at the very edge of the growing part of hisreticula, which is consonant with what we know of proto-plasmic growth in other cases, but not with Mr. Sedgwick'saccount.

But if Mr. Sedgwick can prove that the reticulum is thereand that it grows and spreads far from the nuclei which sub-sequently migrate into it, he must not suppose, as he isapparently so ready to assume, that the inveterate prejudiceof morphologists will prevent their accepting his conclusionsbecause of their theoretical difficulties. If his case is proved,it will be accepted, but he must prove it up to the hilt.

And if he does prove it, what then ? It will be an isolatedcase, of secondary significance: merely another addition toour experience of the very various phenomena displayed in

174 GILBERT C. BOURNE.

organic growth. For thousands of instances point to the factthat normal growth is effected in a very different way, bymitotic division of the nucleus preceding and directing theformation of a discrete or concrescent cell-corpuscle. Therecent researches of cytologists are too many, too good oftheir kind, and too consistent to admit of any other conclusion.

A Criticism of the Cell-Theory; being an Answer to Mr ......A OEITIOISM OF THE GULL-THEORY. 137 A...

Documents

Transcript of A Criticism of the Cell-Theory; being an Answer to Mr ......A OEITIOISM OF THE GULL-THEORY. 137 A...