12 Johnstone, Calajnites (Calamilina) varitvis, Sternb. 



Internode. Measurement in mm 



38 .9- 



)9 22-50 



Incomplete 

 Lowest internode 40 17-' ) P^^od I. 



Total 321-50 mms. 



From this table we see to be true of each period : — 



{a) That the shortest interval always succeeds the 

 branch internode. 



{b) That in the 3rd and 4th periods there are four or 

 five very short internodes just above the branch line. 



(^) That the scar internode itself is shorter, in two 

 out of three cases, than the top one of the preceding 

 period. 



{d) That the top internode of each period is shorter 

 than that below it. 



(<?) That there are irregularities in increase and 

 decrease which may be due to fluctuating conditions of 

 supply. 



Summarising — the minimum length occurs imme- 

 diately above the branches ; there is a gradual increase 

 (usually slower at first) up to the maximum in the last 

 internode but one ; the internode below the row of scars 

 is in two out of three cases less than the maximum, but 

 greater than the branch node; there is an abrupt drop 

 from this last to the minimum. 



Kidston (8) reproduces a fine stem of Catamites 

 SacJisei, Stur. (P/. XIII., Fig. i.) Each of its two rows 

 of branches is supported by a long internode, and one of 

 them shows plainly two short lengths placed above the 

 branches. 



In his paper communicated to the Linnean Society of 

 London in 1909, Mr. Horwood (9) describes two pith 

 casts of Calamites i^Calauiatina) Schiitzei, Stur, and 

 includes for comparison the results of a series of observa- 



