182 



presented a somewhat unusual appearance where they were 

 growing over and around large rocks. The cross-section, in- 

 stead of having the usual circular form, was more or less roughly 

 elliptical, the roots being compressed laterally and expanded ver- 

 tically. Such an appearance at once suggested unequal growth 

 of the secondary layers of the wood and examination proved that 

 this was the case. 



In the specimen brought home and sectioned the greatest 

 horizontal diameter of the root was 24 mm., while the vertical 

 diameter measured 61 mm., exclusive of the rind, which was uni- 

 formly I mm. thick. As a result, the organ presented an almost 

 plate-like form, suggestive of the supporting roots of some trop- 

 ical trees. A section was obtained, thin enough to count the 

 annual rings, and it could be seen that up to about its twenty- 

 fifth year the development of the wood-layers was almost normal, 

 with a slight tendency to epinastic growth (see figures). At this 

 time the root had attained the diameter of about 16.5 mm., so 

 that in breadth its subsequent growth was not more than 8 mm., 

 while vertically it extended five and a half times as much. After 

 the twenty-fifth year the annual rings were to be traced only with 

 great difficulty on the sides, while above and below they were 

 often 2 mm. and sometimes 4 mm. wide. There was not a great 

 deal of difference in the rate of thickening on the upper and lower 

 sides, though the hyponastic growth had a tendency to exceed 

 the epinastic, especially from the thirty-third to forty-second 

 years. The organic center of the root was then not greatly dis- 

 placed from the actual axis of the organ. Such a condition has 

 been described by C. Schimper as diplonasty. About 25 or 30 

 cm. further back, on the root, the hyponastic growth was more 

 pronounced and more irregular. A hasty examination showed 

 that the wood elements were smallest where the rings were com- 

 pres.sed, but there seems also to be some difference in the number 

 of wood-cells present in the different regions, being more, of 

 course, where the ring was widest. This is worthy of notice be- 

 cause, in at least some cases of irregular thickening, it has been 

 stated that the difference in the thickness of the rings was due 

 alone to difference in the size of the wood elements. 



