28 EMBRYOGENESIS IN PLANTS 



mass, and therefore the ratio of internal to surface free energy, thus 

 tends to maintain equihbrium in the growing organism. 



D'Arcy Thompson has argued cogently — though not all observers 

 share his views — that surface forces, in particular surface tension, are 

 important in determining the external shapes of small objects such as 

 embryos, and that they may also account for their characteristic mode 

 of segmentation. A conception of this kind would help to account for 

 the fact that species, which are quite unrelated taxonomically, may 

 nevertheless have comparable segmentation patterns in their embryo- 

 geny. His argument is as follows. If equilibrium is to be maintained in 

 a growing embryo, the free surface energy must be reduced to the 

 minimum, i.e. only minimal surfaces must be created by cell division.^ 

 This conception seems to have an apt application to the segmentation 

 patterns seen in the embryos of plants, Fig. 6. The successive segmen- 

 tations in the embryogeny of a leptosporangiate fern, for example, are 

 in very close conformity with those which we should expect to find in 

 an ideal growing spherical system, subdividing by walls of minimal 

 area, (see D'Arcy Thompson, 1942, pp. 580 et seq). Again, in flowering 

 plants, in which the embryo is immersed on all sides in the liquid contents 

 of the embryo-sac (except at the region of attachment of the suspensor), 

 the initial phase of development consists essentially in the enlargement 

 and highly regular segmentation of a subspherical or club-shaped body. 

 After some time, in relation to other factors, differential growth sets in. 

 In general, the segmentation pattern continues to develop in conformity 

 with D'Arcy Thompson's postulates, but it is modified by the diff'erent 

 rates of growth in different directions, by the inception of locahsed 

 growth centres, and so on. 



In embryos of quite different systematic affinity, closely comparable 

 cellular patterns are found. As we have seen, these remarkable homo- 

 logies of organisation seem to be largely determined by physical factors. 

 In other words, the same physical factors may become incident in 

 different, gene-controlled metabolic systems. It also begins to be 

 understandable how a genie change aff'ecting some particular metabolic 

 process may have an effect on the embryonic pattern. 



^ This principle of cell division by walls of minimal surface has a wide application in the Plant 

 Kingdom. It was Berthold (1886) who, propounding the principle, compared the forms of many 

 cells and the disposition of their dividing walls with those assumed by a system of weightless films, 

 under the influence of surface tension. In the same year, quite independently, Errera definitely 

 ascribed to the embryonic cell wall the properties of a semi-liquid film and deduced that it must be 

 subject to ordinary physical laws and must accordingly assume a form in conformity with the 

 principle of minimal areas. Another conception which we owe to Errera is that the partition wall 

 formed in a dividing cell tends to be such that its area is the least possible by which the given space- 

 content can be enclosed. In an ovoid body with metabolic homogeneity, the partition wall would 

 be median and transverse; but if there were aggregations of different metabolites at the two poles, 

 with concomitant different energy relationships, the partition wall would separate off two cells of 

 equal energy but they would probably be of different sizes. 



