STELAR THEORIES 333 



siphonostele (with an internal endodermis and phloem as well 

 as the normal outer phloem and endodermis) ; the ectophloic 

 siphonostele (with external phloem and endodermis only) ; and 

 finally to the condition obtaining in the majority of higher plants. 



This theory may be illustrated by the consideration of the 

 structure of certain Osmundaceae. 



In Osmunda cinnamomea the mature structure consists of a 

 ring of vascular bundles enclosing a central mass of ground- 

 tissue. Internal phloem, though not general, occurs here and 

 there, more especially below the bifurcation of the stem ; there 

 may also be an internal endodermis. On the other hand, in 

 Osmunda regalis and other plants of the Order, the internal 

 phloem and endodermis are absent. 



Hence, on the whole, the type of structure corresponds to 

 Van Tieghem's medullated monostele ; but, according to Jeffrey, 

 the collateral bundles have arisen from concentric strands of 

 vascular tissue, the inner phloem of which has become obliterated, 

 evidence of this being supplied by the occasional appearance of 

 internal phloem and endodermis in Osmunda cinnamomea. 



This view is disagreed with by Boodle, Chandler, and others, 

 chiefly on developmental grounds; thus Scott (53) remarks that 

 "this theory of reduction is no doubt a tenable one, but as it 

 receives no support from the development of the young plant, 

 its basis is weak." On the other hand, the theory appears to be 

 supported in particular cases, for Jeffrey has described the 

 amphiphloic monostele as giving rise, through reduction, to 

 the medullated monostele. For instance, in Antrophyum semi- 

 costafnm, A. reticidatum, A. plantagineum, Vittaria elongata, 

 Davallia stricta, and Adiantnm pedatum the vascular cylinders 

 have lost entirely, or in part, the internal endodermis and 

 phloem. On the other hand, the steles of Vittaria lineata, 

 Davallia plumosa, and D.fijiensis are quite normal. Thus, in the 

 plants first cited, a strong case may be argued for the medullated 

 monostele being derived from the amphiphloic siphonostele, but 

 it does not follow that the same holds good in all cases. As 

 regards this last point Scott (53) remarks as follows: "An 

 apparently protostelic structure is known to have arisen in 

 certain cases {e.g. water-plants) by reduction, yet no one doubts 

 that in other groups the protostele is really primitive." 



The main features of Jeffrey's position are as follows : 



(1) The endodermis has a morphological value. 



