8 On the Physiology of Bot ant/. 



there is a peculiar leaf quite different from the one that swims, 

 which is made on purpose to cover it, and inclose it from the 

 air ; it generally contains two or three buds, and conveys them 

 safe to the surface of the water ; and when they have passed just 

 above it, the leaf drops. 



There is also another beautiful fact which still lends its aid to 

 the passing up of the flower on those plants that have leaves and 

 flowers on the same stem : their roots are perfectly circular (see 

 fig. 8) ; but as soon as the flower is to pass on, and be carried 

 up to where it will issue from the bosom of the leaf, the stem 

 becomes either square, pentagonal, hexagonal, or gains some 

 -form that will enable it in the interstices of these projections, to 

 pass the flowers up, and convey them where they will issue from 

 the axilla of the leaf, without pressure or difficulty ; particularly 

 in the Pentandria digynia plants. In the Oeiianthe, in the 

 Heracliinn spondylium (fig, 9), the Atriplex (fig. 10), I have 

 often s>een when the stem has been cut with a sharp knife, and 

 the plant then laid on the table a few hours, the flowers have ap- 

 peared near the tenth of an inch above the edge of the stem, 

 having lengthened within that time: and often, if closely watched, 

 and the stem is very clear in the bark, the leaves may be seen 

 at the interior. 



In herbaceous plants when above the root, they have rarely 

 buds : but a curious fact is observed when the flowers have at- 

 tained their proper situation, and risen to the axilla of the leaves, 

 the figure of the stem sometimes wholly alters, and part of the 

 stem will show, by the outward skin falling in, that a large aper- 

 ture was allowed for the passing up of the buds ; — for the grooves 

 will afterwards exhibit a very different appearance (see fig. 11, 

 which was fig. 9). The Datura also totally alters its shape, 

 losing almost one-third of the stem. In the Sawbiicus, which i» 

 hexagonal, one deep division is principally bent in, and sinks 

 after the flowers have passed. I think therefore I need not press 

 the evidence further. What I have already advanced would, I 

 should conceive, convince any one. This will prove two of the 

 laws I wish to establish. 1st. That the root is the laboratory of a 

 plant ; 2d. That the flower-bud is formed in the root. 



In my next I shall show the third law : " That the heart or 

 embryo of the sesd is formed in the radical or lowest part of 

 the root, but does not join the seed till it enters the seed-vessel 

 for the purpose." That the heart of the seed should be formed 

 in the root, cannot be such a wonder ; when the flower-bud is 

 protruded there, and when the embryo containing that shoot 

 which forms the first growth of the next year is there taken in, 

 where they are alone to be found : from hence they pass up the 

 alburnum, and into the seed-vessel. 



I am 



