148 NATURAL SCIENCE. September, 



walls. He often fails to find this endodermis when he cuts his sections 

 of the stem ; but then sections do not always turn out as they should. 

 He goes on bravely, but schizostely puzzles him, and when he comes 

 to EquiseUim he gives it up as hopeless. We would advise him to 

 put a pin through those pages in his text-book, and go to some library 

 where he can get hold of the April and May numbers of Science Progress 

 (it is much too expensive for him to think of buying — an opinion 

 which, unfortunately, is shared by many librarians), and read an 

 account of the history of the theory, with criticisms thereon by 

 Mr. A. G. Tansley. Meanwhile, we will put him up to one or two 

 points. Plausible though it seem, there is really no connection 

 between the dermatogen, periblem and plerome business, and the 

 stelar theory of Van Tieghem. Perhaps, though, relying on his text- 

 book, he does not even know it as Van Tieghem's theory. If he goes 

 into the question, he will find that recent observers are not by any 

 means certain about these three initial layers at the growing point, 

 and that the stelar theory must, in the present state of our knowledge, 

 be considered quite apart from them. Briefly, it is something like 

 this. There is a striking similarity in the internal structure of all 

 roots, as Van Tieghem showed nearly five-and-twenty years ago. 

 They contain a central cylinder, in which, before matters become 

 complicated by secondary thickening, we find strands of wood and 

 bast separated by parenchymatous " conjunctive cells," the whole 

 surrounded by a pericavihium, or pericycle, and bounded on the outside 

 by the innermost layer of the cortex, which shows the characters that 

 we associate with an endodermis. The great service of Van Tieghem 

 to the study of plant anatomy was in showing that this arrangement 

 in the root is continued through the hypocotyl into the stem, and 

 that in stems of many seed-plants it is possible and easy to trace a 

 pericycle continuous with that of the root and surrounding the system 

 of vascular bundles separated by conjunctive tissue, with which is 

 included the pith. Continuous also with the endodermis of the root 

 is a similar layer, with similar characteristics, forming the limit of the 

 cortex towards the central cylinder or stele. The root, hypocotyl and 

 stem have one continuous stele — are, in fact, vwmstelic. There are cases, 

 however, where the very best of sections reveal no endodermal layer 

 round the central bundle-system. Now, an endodermis cannot be 

 imagined. As Strasburger has pointed out, it is a layer whose cells 

 have certain histological characters to which we have referred, these 

 characters being associated with certain physiological properties. It 

 is, in fact, a layer that will transmit water, but not air ; as such it is 

 eminently of service in enclosing the water-conducting system, and 

 therefore very frequently found surrounding it ; but it may, and does, 

 occur elsewhere, and must not be considered as an integral part of 

 the stelar theory. To avoid this Strasburger suggests a new term, 

 phlocotevma, for the layer next outside the stele and surrounding it. 

 The phloeoterma may have endodermoid characters, or it mayba 



