224 THE SCOTTISH BOTANICAL REVIEW 



the embryo and development of embryo-sac are affected by (i) space 

 rotations, (2) mode of food-supply, (3) shape of embryo, and depend 

 on future form of seedling and mode of freeing itself from the seed- 

 tissue. The embryo-sac is long and narrow, and bent upon itselfj 

 evidently due to nature of surroundings. The suspensor conveys 

 food. The cotyledons are absorptive, and endosperm prepares and 

 draws upon the store. Is the function of the cotyledons to absorb 

 food and pass it on ? The embryo exhibits adaptation to present, 

 past, and future environments. Wide embryo-sacs have a lateral 

 cotyledon, and narrow embryo-sacs, as in Alisma, an apparently 

 terminal cotyledon. 



If the single cotyledon was lateral, was that of the primitive 

 Angiosperm lateral? It is possible to imagine a lateral member 

 split into two, and that one member has revolved through i8o\ If 

 it was dicotylar, the monocotylous form might be derived by 

 suppression of one or fusion of two members. The cotyledon was 

 single in pseudomonocotyledons, and derived from two presumably. 

 The development is like that of Tamus. Sterckx found in 

 Ranunculus ficaria a bilobed venation in the cotyledon which 

 suggested fusion of two cotyledons. Analogy with Gymnosperms 

 favours this, but development of the embryo-sac yields no evidence, 

 its shape being due to environment. 



As regards embryology after germination, certain types of mono- 

 cotylous plants have prmiitive vascular tissue. In dicotyledons the 

 tetrarch type is found as well as the diarch type. 



In Nigella it is diarch, but there is transition to a tetrarch root. 

 In Althea. which is tetrarch, there is an approach to diarch structure, 

 like Liriodendron. The oldest Gymnosperms exhibit tetrarch struc- 

 ture and also diarch structure. In Pteridosperms probably the 

 structure was tetrarch. Thus presumably the primitive forms 

 were either tetrarch or diarch. In monocotyledons the structure 

 is more variable, but there is a series of extreme types and also 

 intermediates. 



The common origin indicates presumably a primitive type. Such 

 are Liliaceae, e.g. Anemarrhena, which is tetrarch, and they are con- 

 nected with others, e.g. Iridacese, Aroideje, whilst it is in fact a radicle 

 of Liliifiorae, and hence may be of monocotyledons in general, recapi- 

 tulating ontogenetically the history of the group. There is a strong 

 resemblance between the structure of Anemarrhena and the dicotylar 

 Althea, which is near the primitive gymnospermous type, and prob- 

 ably primitive. Hence probably the primitive Angiosperm type 

 had a tetrarch stem-structure. Would this necessarily be correlated 

 with two cotyledons? Dicotylar forms with cotyledons each sym- 

 metrical about the median plane are symmetrical about two vertical 

 planes. There is no symmetry about one plane, where reduction 

 occurs. In monocotylous stems receiving traces from a single lateral 

 there is only symmetry about the median plane. The primitive 

 Anemarrhena, Galtotiia, etc., are exceptions. That the tetrarch 

 skeleton symmetrical about two planes is found in these suggests 



