302 ON THE INTERNAL FORM [ch. 



and which we now know to be not inherent in the nature of proto- 

 plasm nor of Hving matter in general, but to be due to various 

 causes, natural as well as artificial*. The microscopic honeycomb 

 structure of cast metal under various conditions of coohng is an 

 example of similar surface-tension phenomena. 



Such then, in briefest outhne, is the typical structure commonly 

 ascribed to a cell when its latent energies are about to manifest 

 themselves in the phenomenon of cell-division. The account is 

 based on observation not of the hving cell but of the dead : on the 

 assumption, that is to say, that fixed and stained material gives a 

 true picture of reahty. But in Robert Chambers's method of micro- 

 dissection f, the hving cell is manipulated with fine glass needles 

 under a high magnification, and shews us many interesting things. 

 Chambers assures us that the spindle fibres never make their 

 appearance as visible structures until coagulation has set in; and 

 that astral rays are, or appear to be, channels in which the more 

 fluid content of the cell flows towards a centrosome:|:. Within the 

 bounds to which we are at present keeping, these things are of no 

 great moment; for whether the spindle appear early or late, it still 

 bears witness to the fact that matter has arranged itself along 

 bipolar fines of force; and even if the astral rays be only streams 

 or currents, on lines of force they still approximately he. Yet the 

 change from the old story to the new is important, and may make 

 a world of diff'erence when we attempt to define the forces concerned. 

 All our descriptions, all our interpretations, are bound to be 

 influenced by our conception of the mechanism before us; and he 



* Arrhenius, in describing a typical colloid precipitate, does so in terms that 

 are very closely applicable to the ordinary microscopic appearance of the protoplasm 

 of the cell. The precipitate consists, he says, "en un reseau d'une substance solide 

 contenant peu d'eau, dans les mailles duquel est inclus un fluide contenant un peu 

 de colloide dans beaucoup d'eau. . . . Evidemment cette structure se forme a cause 

 de la petite difference de poids specifique des deux phases, et de la consistance 

 gluante des particules separees, qui s'attachent en forme de reseau " {Rev. Scientifique, 

 Feb. 1911). This, however, is far from being the whole story: cf. (e.g.) S. C. 

 Bradford, On the theory of gels, Biochem. Journ. xvii, p. 230, 1925; W. Seifritz, 

 The alveolar structure of protoplasm, Protoplasma, ix, p. 198, 1930; and A. Frey- 

 Wissling, Submikroskopische Morphologie des Protoplasmas, Berlin, 1938. 



t See R. Chambers, An apparatus. . .for the dissection and injection of living 

 cells, Anatom. Record, xxiv, 19 pp., 1922. 



X This centripetal flow of fluid was announced by Biitschli in his early papers, 

 and confirmed by Rhumbler, though attributed to another cause. 



