INTRODUCTION. XX'xiii 



present in Neillia, and in others traces are to be seen adherino- to 

 the inner membrane of the testa. — The texture of the albumen is 

 frequently consulted with advantage : in all Rubiacese it is horny 

 or fleshy ; Euphorbiaceae, oily ; Grasses, Polygoneae, Chenopodeee, 

 mealy ; in Annonaceae, it is ruminated, &c. ; but among Apocynese, 

 which have solid albumen, it is ruminated in Alyxia. 



The direction of the Embryo within the testa, which is indi- 

 cated in the ovulum by the foramen, is one of the very few characters 

 to which we know of no exceptions; and if it were a less obscure 

 point of structure, it would consequently be one of the most useful. 

 For example, in all Cistineae, UrticeaB, and Polygoneae, the radicle 

 is not turned towards the hilum, as in other tribes, but takes an 

 opposite direction ; and these orders are distinguished from their 

 allies by this, better than by any other known character. 



The number of Cotyledons is generally believed to be one of 

 the most important means of distinguishing the great natural divi- 

 sions called Monocotyledons, Dicotyledons, and Acotyledons ; and it 

 is a most curious fact, that this goes along with anatomical structure. 

 There are, however, plants among Monocotyledons with two coty- 

 ledons, as the common Wheat ; and among Dicotyledons with only 

 one, as Pensea and some Myrtacese ; or even none, as Cuscuta and 

 Utricularia ; or several, as Schizopetalon in Cruciferse, Benthamia 

 in Boragineae, Ceratophylleae, and most Coniferae. — To the relative 

 position of the cotyledons there are not the same objections, v/hence 

 the character of Dicotyledons has been found to consist in the coty- 

 ledons being opposite to each other ; of Monocotyledons, in their 

 being alternate with each other, if there is more than one ; and of 

 Acotyledons, in germination taking place from no particular point, 

 rather than in their number. 



The only remaining character of vegetation which I find it 

 necessary to notice is a singular and very uncommon one, which 

 distinguishes a few small families of plants. This consists in the 

 presence of the remains of the Amnios around the embryo in its 

 perfect state : the amnios always surrounds the embryo in an early 

 state, but is most commonly absorbed before the formation of the 

 embryo is completed; but in Saururea^, Piperaceae, and Nymphreaceae, 

 its remains surround the embryo in the form of a sac, which was 

 mistaken by Richard, who did not understand its nature, for a 

 peculiar appendage of the embryo, or rather for a particular form 

 of the radicle, — an hypothesis which that distinguished botanist 

 supported with great skill, but which is now generally abandoned. 



I have now gone through the whole of the characters of which 

 botanists make use in distinguishing and determining the affinities 

 of plants, and I think it must be apparent that the difficulties 



