238 EMBRYOGENESIS IN PLANTS 



begins to bend as it adapts itself to the curved embryo sac, Fig. 54 

 {see also Fig. 83). 



In an investigation of the late embryogeny in Pisimi, Reeve (1948) 

 has shown that the epicotyl apex is initiated by periclinal divisions in 

 surface cells before the cotyledons are formed. During the early coty- 

 ledonary development a simple apical dome, with one or two tunica 

 layers and a shallow corpus, can be observed. At first the tunica 

 layers undergo frequent periclinal divisions and the apex enlarges; in 

 the fully developed embryo a well organised apical meristem of two 

 layers is present. In seedling and adult plants this apex becomes more 

 highly arched with three or four tunica layers, but its general organisa- 

 tion is as in the embryo. A rib meristem, which is first apparent in the 

 meristematic tissue of the embryo axis, is important in that it contributes 

 to the early rapid growth of the cotyledons. In the young cotyledons 

 the distal meristem is not a highly organised one, the cell divisions being 

 apparently at random. Root formation lags behind that of the shoot, 

 but the first columella tiers, from which the generative root meristem is 

 formed, can be distinguished about the same time as the epicotyl 

 initials. This kind of study is important because it emphasises the need 

 for a full re-examination of histogenesis in seed plants, beginning with 

 the young embryo. Here the reader may also be referred to a develop- 

 mental study of the apical meristem in different varieties of Avena 

 sativa at different temperatures by Hamilton (1948). 



A critical analysis of the several embryonomic types will call for a 

 study of the factors which determine the specific distribution of growth. 

 The histological pattern is probably indicative of differential metabolic 

 activity in the embryo, cell division, according to D'Arcy Thompson 

 (1917), tending to restore equilibrium in the system. In any species it is 

 reasonable to suppose that the particular segmentation pattern is gene- 

 determined, for the different kinds, or intensities, of metabolic activity 

 in different parts of the embryo are ultimately to be referred to factors in 

 the hereditary constitution. But in that different reaction systems may 

 yield closely comparable embryonic cell patterns, the use of embryo- 

 logical data in taxonomy and other comparative studies should be 

 viewed with reserve in the present state of knowledge. 



In the early stages there are no essential anatomical differences 

 between the embryos of monocotyledons and dicotyledons. Fig. 

 56F-K shows the development of a simple monocotyledonous embryo, 

 that o[ Luzula. The terminal cell divides longitudinally, as in Capsella, 

 and the basal cell transversely. A further longitudinal division in the 

 terminal cell at right-angles to the first yields a quadrant. The subjacent 

 cell also divides by a longitudinal wall. Later, as Fig. 56j shows, the 

 embryo becomes spherical, the upper cell-products of the basal cell 



