STRUCTURE.] DUCTS MOHI/S VIEWS. 85 



however, furnished me with a perfectly satisfactory result, 

 and my former researches on the roots of Palms and other 

 monocotyledonous plants having shown me the greater faci- 

 lity of studying the development in this organ than in the 

 trunk, I submitted the roots of Tradescantia to a very atten- 

 tive examination, the results of which I consider to be quite 

 conclusive. The examination of the roots presents this great 

 advantage over that of the stems, that in the larger vessels, 

 placed nearer to the centre, the fibres are not developed until 

 a sufficiently late period, when their longitudinal growth is 

 already terminated. At the period when the fibres of the 

 vascular utricles are developed, these utricles have not only 

 already attained to a considerable size, but the fibres in them 

 are also, from the beginning, arranged at greater distances 

 from each other, and their successive development may be 

 followed in detail, step by step, from one end of the root to 

 the other. This examination is rendered easier in conse- 

 quence of the vessels being deposited in a very transparent 

 cellular tissue. In these researches I have recognised with 

 the greatest clearness, and with a perfect conformity to what 

 I had previously observed in the roots of Palms, that, from 

 the time when the fibres make their appearance, and when 

 they are still so tender, narrow, and transparent, that it is 

 often only possible to see them with a faint light, they 

 already present all the different modifications of form which 

 are observed in the perfect vessels. We then find, as at a 

 later period, the same alternation of annular, and spiral, and 

 reticulated fibres ; but I have never seen the least trace of 

 the formation in all vascular utricles of a spiral fibre whose 

 coils would unite in pairs, and the portions of the spiral fibre 

 serving as the means of union be absorbed ; and I consider it 

 as perfectly impossible that this transition of spiral vessels 

 into annular vessels, if it existed, could have escaped me, 

 because in a great number of roots I have followed the 

 vessels from the moment when the utricles presented closed 

 cells with thin parietes, and inclosed a nucleus. 



"Hence it results that the development of the annu- 

 lar vessels agrees with the observations made on the per- 

 fect vessels. Researches into these two organs show that 



