APPENDIX. Xo. V. 481 



sion of these, recourse must be liad to natural causes, or such as are uncon- 

 nected with human agencj-. But the necessity of callirjg in the operation of 

 these causes inipUes the adoption of that theory according to which each 

 species of plants is originally produced in one spot only, from which it is gra- 

 dually propagated. Whether this be the only, or the most probable opinion 

 that can be held, it is not my intention to enquire : it may however be stated 

 as not unfavourable to it, tliat, of the Dicotyledonous plants of die Usts, a con- 

 siderable number have the embryo of the seed highly de\ eloped, and at the 

 same time well protected by tlie texture of its integuments. 



This is the case in Malvaceae, Convolvulaceae, and particularly in Legumi- 

 nosae, which is also the most numerous family in the lists, and in several of 

 whose species, as Guilaudlna Bonduc, and Abnis precatorius, the two conditions 

 of developement and protection of the embryo coexist in so remai-kable a de- 

 gree, that I have no doubt the seeds of tliesc plants would retain their vitality 

 for a great length of time, either in the currents of the ocean,* or in the di- 

 gestive organs of birds and other animals ; the only means apparently by 

 which tlieir transportation from one continent to another can be efiected : and 

 it is deserving of notice that these seem to be the two most general plants 

 on the shores of all equinoctial countries. 



The Dicotyledonous plants in the lists wliich belong to other famihes have 

 die embryo of the seed apparently less advanced, but yet in a state of consider- 

 able developement, indicated either by the entire want or scanty remains of 

 albumen: the only exception to this being Leea, in which tlie embryo is 

 many times exceeded in size by the albumen. 



In the Monocotyledonous plants, on the other hand, consisting of Gramlneae, 

 Cyperaceas, Gloriosa, Flagellaria and Pistia, the embryo bears a very small 

 proportion to the mass of the seed, which is formed of albumen, generally 

 farinaceous. But it may here be observed diat the existence of a copious 

 albumen in Monocotyledones does not equally imply an inferior degree of 



* Sir Joseph Bnnks informs me, that he received some years ago the drawing of a 

 jilant, which his correspondent assured him was raised i'rora a seed found on the west 

 toast of Ireland, and that tile plant was indispulMy Guilandina Bonduc. Linnsusalso 

 kcems to have been acquainted with other instances of germination having taken plac; 

 in seeds thrown on shore on tlie coast of Norway, /'irfd Coloniee Plantarum, p. 3, in 

 Amcen. Acad. vol. 8. 



3Q 



