August 27, 1891] 



NATURE 



411 



sharp curve close to the tip of a geotropic root, and the long 

 gradual curve of an apogeotropic shoot, are necessary conse- 

 quences from the manner in which growth is distributed in these 

 parts. He demonstrated that rectilinear growth and geotropic 

 curvature require the same external conditions ; that, for in- 

 stance, a temperature low enough to check growth also puts a 

 slop to geotropism. 



The distribution of longitudinal growth, which produces geo- 

 tropism, was afterwards studied by Sachs {Arbeiten, i. p. 193, 

 June 1S71), who thoroughly established the fact that the convex 

 side grows faster, while the concave side grows slower, than if 

 the organ had remained vertical and uncurved. 



These facts are of interest in themselves, but they do not, 

 any more than Frank's results, touch the root of the matter. 

 Until we know something of the mechanics of rectilinear 

 growth, we cannot expect to understand curves produced by 

 ^'lowth. The next advance in our knowledge did, in fact, 

 ..ccompany advancing knowledge of rectilinear growth. It 

 began to be established, through Sachs's work, that turgescence 

 is a necessary condition of growth. A turgescent cell is one 

 which is, as it were, over-filled with cell sap"; its cell-walls are 

 stretched by the hydrostatic pressure existing within. In 

 osmosis, which gives the forre by which the cells are stretched, 

 a force was at hand by which growth could be conceived to be 

 caused. The first clear definition of turgor, and a sta'ement of 

 its importance for growth, occurs in Sachs's classical paper on 

 growth {Arbeiten, p. 104, August 1871). 



As soon as the importance of turgor in relation to growth 

 w as clearly put forward, it was natural that its equal importance 

 wiih regard to growth-curvatures should come to the lore, and 

 that increased growth on the convex side (leading to curvature) 

 should be put down to increased internal cell-pressure in those 

 tissues. In the fourth edition of Sachs's " Lehrbuch " (1874^, 

 l.ng. trans., 1882, p. 834, such a view is tentatively given, but the 

 author saw very clearly that much more evidence was needed 

 before anything like a conclusion as to the mechanism of move- 

 ment could be arrived at. The difficulty which faced him was 

 not a new one— in a slightly different form it had occurred to 

 Hofmeister— the question, namely, whether the curvatures of 

 acellular an I multicellular organs depend on the same or on 

 different causes. If one explanation is applicable to both, then 

 we must give up as a primary cause any changes in the osmotic 

 force of the cells. For no change in the pressure inside a cell 

 \\\\\ produce a curvniure in that cell, whereas, in a multicellular 

 organ, if in the cells in one longitudinal half an increase of 

 osmotic substances takes place, so that the cell-walls are subject 

 to ni eater stretching force, curvature will lake place. 



On the other hand, if the cause of bending of acellular and 

 multicellular organs is the same, we must believe that the curva- 

 ture takes its origin in changes in the cell-walls. In an acellular 

 organ, if the cell-membranes yield symmetrically to internal 

 pre-sure, growth will be in a straight line ; if it yields asym- 

 metrically it will curve. Thus, if the membrane along one side 

 of a cell becomes more or less resisting than the rest of the 

 membrane, a curvature will result. 



If we are to apply strictly the same principle to acellular and 

 multicellular organs, we must suppose that the whole organ 

 curves, because each individual cell behaves like one of the 

 above-described free cells, the curvature of the whole resulting 

 from the sum of the curves of the separate cells. This was 

 Frank's view, and it also occurs in Sachs's "Text-book " (1874), 

 Eng. trans., 1882, p. 842. 



Are we bound to believe that the mechanism of acellular and 

 multicellular curvalures is so strictly identical as Frank sup- 

 posed? In the first place, it is not clear why there should be 

 identity of mechanism in the movements of organs or plants of 

 completely different types of structure. The upholders of the 

 identity chiefly confine themselves to asseveration that a common 

 explanation must apply to both cases. I believe that light may 

 be thrown on the matter by considering turgescence, not in 

 relation to growth, but in regard to stability of structure. 



An acellular organ, such as the stalk of the sporangium of 

 Mucor, owes its strength and stiffness to the tension between the 

 cell contents and the elastic cell-wall, but it does not follow from 

 this I hat in multicellular organs strength and stiffness are due to 

 the sum of the strength of its individual cells. Indeed, we know 

 that it is not so : the strength of a multicellular organ depends 

 on the tension between pith and cortex. It is, in fact, a model 

 of the single cell ; the pith represents the cell-sap, the cortex 

 the cell-wall. Here, then, it is clear that the function performed 



NO. 1 139, VOL. 44] 



by the cell-wall in one case is carried out ly cortical tissues in 

 the other. If this is the case for one function, there is no 

 reason why ii should not hold gocd in another, viz. the machinery 

 of movement. 



If we hold this view that the cortex in one case is analogous 

 with a simple membrane in the other, we shall not translate the 

 unity of acellular and multicellular organs so strictly as did 

 Frank. Indeed, we may fairly consider it harmonious with our 

 knowledge in other departments to find similar functions per- 

 formed by morphologically different parts. The cortex of a 

 geotropic shoot would thus be analogous with the membrane of 

 I a geotropic cell in regard to movement, just as we know that 

 j these parts are analogous in regard to stability. 

 I In spite of the difficulties sketched above, one writer of the 

 first rank, namely, H. de Vries, has upheld the view that growth- 

 curvatures in multicellular organs (j^£?/. Zeitung, 1879, p. 835), 

 are due to increased cell-pressure on the convex side ; the rise m 

 hydrostatic pressure being put down to increase of osmotic sub- 

 stances in the cell-sap of the tissues in question. Such a theory 

 flowed naturally from De Vries's interesting plasmolytic work 

 {ibid. 1877, p. i). He had shown that those sections of a 

 turgescent shoot which were in most rapid growth /show the 

 greatest amount of shortening when turgescence is removed by plas- 

 molysis. This w as supposed to show that growth is proportional to 

 the stretching or elongation of the cell-walls by turgor. Growth, 

 according to this view, consists of two processes : (l) of a tem- 

 porary elongation due to turgescence, and (2) of a fixing process 

 by which the elongation is rendered permanent. De N'ries 

 assumed that where the elongation occurred, its amount must be 

 proportional to the osmotic activity of the cell contents ; thus 

 neglecting the other factor in the problem — namely, the vari- 

 ability in the resistance of the membranes. He applied the 

 plasmolytic method to growth-curvatures, and made the same 

 deductions. He found that a curved organ shows a flatter 

 curve ^ after being plasmolyzed. This, according to his previous 

 argument, shows that the cell-sap on the convex is more power- 

 fully osmotic than that on the concave side. This again leads to 

 increased cell-stretching, and finally to increased growth. 



The most serious objection to De Vries's views is that the 

 convex half of a curving organ does not contain a greater amount 

 of osmotically active substance." It must, however, be noted in 

 the heliotropic and geotropic curvature of pulvini, there is an 

 osmotic difference between the two halves-* — so that, if the 

 argument Irom uniformity is used against De Vries (in the 

 matter of acellular and multicellular organs), it may fairly be 

 used in his favour as regards the comparison of curvatures pro- 

 duced with and without pulvini. 



It is not easy to determine the extent to which De Vries's views 

 on the mechanics of growth-curvature were accepted. 'i'he 

 point, however, is of no great importance, for the current of 

 conviction soon began to run in an opposite direction.* 



Sachs (" Lehrbuch," ed. 4, Eng. trans, p. 835) had already 

 pointed out that attention should be directea to changes in 

 extensibility of cell-walls as an important factor in the problem. 



Wiesner, in his " Heliotropische Erscheinungen " {lliemr 

 Sitzuiigsb., vol. Ixxxi., 1880, p. 7 ; also in the Denkschrijten, 

 1882), held that the curvature of multicellular organs is due both 

 to an increase of osmotic force on the convex side, and to in- 

 creased ductility'' of the membranes of the same part. He 

 repeated De Vries's plasmolytic experiments, and made out the 

 curious fact that in many cases the curvature is increased instead 

 of being diminished. He attributed the result to the concave 

 tissues being more perfectly elastic than ductile convex tissues, 

 so that when turgescence is removed, the more elastic tissues 

 shorten most, and, by diminishing the length of the concave side, 

 increase the curvature. 



Strasburger, in his " Zellhiiute " (1882), suggested that growth- 

 curvatures are due to increased ductility ot the convex mem- 

 branes, and gave a number of instances to prove that a change 

 to a ductile condition does occur in other physiological procetse;, 

 such as the stretching of the cellulose ring in Ui logonium to a 



' Frank made similar experiments, but failed to find any diminution cf 

 curvature. 



2 Kraus, Ahhand. Kat. Cesell zh Halle, xv., 1882. See also a different 

 proof by VVortmann, Deutsch Bot. Gesell., 1887, p. 459. 



3 Hilburg in Pfeffer's Tubingen. UntersHch., vol. i., 1881, p. 31. 



•• An opportunity will occur later on for referring; to some details of De 

 Vries's work not yet n Jticed. 



5 Weinzierl, Siiziingsb U'ien.. 1877, showed thatstripsof epidermis taken 

 cff the convex side of helio:rop>cally curved flower-stalks of tulip and 

 hyacinth were about twice hs extensible when stretched by a small weight, 

 75 grammes, as approximately c >rrei|)onding strips for the concave side. 



