COTYLEDONS. 613 



not until this has occurred do they unfold and become gi-cen. In tliis penetration 

 of the earth the cotyledons are exposed to so many dangers that special arrange- 

 ments are frequently to be found with a view to protecting their advancing points. 



As the cotyledons push through the ground, a pressure is exercised upon the 

 layei-s of soil above by the straightening of their stalks. The cotyledons raise 

 portions of soil on their backs, so to speak, without actually bursting or boring 

 througli them. In this way the danger of injury is at any rate much diiniiiished, 

 and the supposition is fully justified that cotj'ledons which develop after the type 

 of the Gourd or Asafoetida are those which occur most frequently. Plants whose 

 straight embryo has to pierce through the fruit-husk and the layer of earth above 

 it by means of their conically-folded cotyledon apices are, as already stated, less 

 common. Fig. 145^ shows this rare foi-m in Cardopatium corymbosum. It has 

 also been observed in many other species allied to this composite, and in tlie 

 Mediterranean Atractylis cancellata. 



In all those cases where the cotyledons are withdrawn through a cleft or hole 

 in the seed-coat it seems quite obvious that the aperture should have a diameter 

 at least equal to that of the organs to be withdrawn. As a rule this is so; but 

 occasionally the cotyledons ai'e actually broader than the cleft, and one asks in 

 astonishment how the withdrawal could have been accomplished without injury to 

 their fabric. The feat is performed in the following way. Before the strain on 

 the cotyledons comes, they are folded together, and are then drawn out a.s a long 

 roll through the narrow opening of the integument. Scarcely have they been 

 liberated ere they begin to unroll and spread themselves out flat. This is the case, 

 for example, in the Immortelle {Helichrysum annuum) (see figs. 145^ and 145*), also 

 in the umbellifer, Smyrnium Olusatrum, and in many others. In some plants, e.g. 

 in the Beech (Fagus sylvatica), the cotj^ledons, as long as they remain within the 

 husk, are folded together lengthwise like a fan, and in this position occupy but a 

 limited space. They are also withdrawn from the nut thi-ough a comparatively 

 small aperture, and then expand in a very short time after they are free (see figs. 

 148 ^•-•^). The two cotyledons of Pterocarya caucasica are each divided into 

 four lobes, and each pair of lobes lying close to one another are imbedded in a special 

 excavation in the seed. Altogether the fruit presents four compartments, in each 

 one of which lies such a pair of narrow, closely-packed lobes. The aperture of the 

 nut-like pericarp now affords sufficient space for each pair of folded lobes to be 

 drawn out, nor does their withdrawal occur simultaneously, bait ratlier so that 

 the pairs of lobes emerge one after the other. The cotyledons of Schizopetalon 

 Walkeri behave in a similar manner, each of them being divided into two long 

 nan-ow lobes which are drawn out successively through the small aperture of the 

 spherical seed. Moreover, in the embryos of Pinus there are five or more whorled 

 linear cotyledons (see fig. 148^). These al.so leave the cavity of the seed one after 

 the other. Speaking generally, it would seem that the dimensions and form of the 

 cotyledons are correlated with the shape of the seed-coat or other investment, and 

 with its manner of opening. 



