ii4 Mr. Thomas Hick on 



after secondary thickening has set in.* Whether it is ulti- 

 mately thrown off as in recent roots is not actually proved 

 by my preparations, but one of them suggests that this is 

 the case. Here the cortex is reduced to a relatively narrow- 

 zone, only slightly thicker than that which represents the 

 pericycle and endodermis, and is composed of somewhat 

 similar cells. At the periphery there are indications of a 

 separation layer of periderm, but these are too vague to 

 justify a more definite statement. 



Reviewing the whole of the characters presented by the 

 various forms of Kaloxylon Hookeri, Will., they seem to me 

 to fully warrant the conclusion that the fossils so named are 

 all roots, in the morphological sense at least, and that the 

 three types are merely different stages of development, 

 Whether every root passed through all the three stages, or 

 whether some of them remained permanently in the primary 

 condition, it is impossible to say, but this point is of secon- 

 dary importance. A more pressing question is whether it 

 is possible to connect these roots with any of the stems 

 which are found associated with them in the Lower Coal 

 Measures of Yorkshire and Lancashire. In what follows 

 an attempt is made to answer this question in the affirma- 

 tive, and to advance reasons for regarding Kaloxylon Hookeri y 

 Will., as the root of Lyginodendron Oldhamium, Will. 



Among the details of structure which distinguish Kal- 

 oxylon Hookeri from most other Carboniferous plants are 

 the markings on the walls of the tracheae of the primary 

 and secondary xylem. As already mentioned, these ele- 

 ments are of the pitted type, and the type is so constant that 

 in his Vllth Memoir Williamson tells us that at the time 

 of writing he had not discovered in any of his sections a 

 single barred or spiral vessel. Without asserting that all 

 other kinds of tracheae are altogether absent, the fact that 

 the pitted form is the prevailing one prompts us to look for 



•Fig. i. c. 



