416 Progress of Natural History^ 



evaporation of the internal fluid, and returns again, when the 

 valves are plunged into water before they are entirely dry. 

 According to M. Dutrochet, after incomplete evaporation they 

 still contain a dense fluid, and endosmosis will take place; but 

 after entire desiccation, the water only effects common ab- 

 sorption. If we plunge these same valves of the jBalsamina 

 into a denser fluid than that which they contain, syrup for 

 instance, exosmosis takes place, they immediately cease to 

 curve inwards, and soon turn back, because their larger ex- 

 terior vessels lose more of their fluid than those of the interior. 

 By applying even the little that has been already said, and 

 which does not contain more than half of the details entered 

 into by M. Cuvier, it will be seen that this power acts on the 

 direction of the stem and radicle of the embryo in germinating 

 seeds, even on the tendency of stems and roots to ascend and 

 descend, and also on the curvature of stems, which are upright 

 as long as their proper equilibrium is retained. 



The structure and developements of the vegetable ovulum, 

 which have for many years excited the attention and study of 

 the first botanists in Europe, and which, after so many scien- 

 tific observers, might be thought exhausted subjects, have 

 again been the object of M. de Mirbel's researches, who in 

 some respects confirms, but in others contradicts, the asser- 

 tions of his learned predecessors. He considers the ovulum 

 from its birth, follows it through all its developements, even 

 through its changes of position and exterior forms, by which 

 he has been led to divide seeds into three classes : first, the 

 " Orthotropes," which retain the position they had in the first 

 instance, and which is that of the base exactly opposite to the 

 summit; secondly, the " Anatropes," which change their 

 position so completely as to be reversed; and, thirdly, the 

 " Campulitropes," which curve in an arch or circle, so that the 

 summit approaches the base. A long time ago M. de Mirbel 

 observed, that in square stems with opposed leaves, there are 

 four vascular and ligneous bundles, which correspond with the 

 four angles, and that at the extreme point of insertion of each 

 pair of leaves, these bundles communicate with each other by 

 means of lateral ramifications, which form an annular pad 

 round the stem. The trunk of an old Calycanthus floridus 

 has furnished the author with a fresh confirmation of this fact. 

 The four vascular bundles at the angles of this Calycanthus 

 have grown with the stem, and formed four projections out- 

 side, which appear like cords the size of the fourth finger. 

 Each of these has a cortical covering of its own, ligneous 

 layers placed one above another, large vessels in a circular 

 series in the wood, radiations which proceed from the centre 



