18 T. G. HALLE, LOWER DEVONIAN PLANTS FROM RÖRAGEN IN NORWAY. 



specimens which do not show the spines themselves but only the scars marking their 

 position: such specimens are shown in pl. 1, figs. 21 — 23, and pl. 2, figs. 1 — 5. All the 

 specimens represent stems which occasionally branch and usually are preserved as im- 

 pressions and therefore appear flattened. That their original shape was cylindrical is 

 evident among other things from the thickness of the few specimens in which the original 

 tissue is preserved in a carbonized state or has been replaced by some ferruginous matter: 

 such specimens retain something of the original shape. The diameter of the stems is 

 fairly uniform; the thickest specimens in which spines or scars can be recognised with 

 certainty hardly exceed 10 mm. in breadth on the impressions, and the average breadth 

 is about 5 mm. The narrowest branches seen in connection with specimens of this 

 kind and still retaining the leaves or spines, or showing scars of them, measure about 



2 mm. in breadth. 



The branching is very characteristic: it is always dichotomic, and the two sister- 

 branches are mostly equally developed. Such is seen to be the case for instance in the 

 specimen in pl. 1, fig. 21, which is twice bifurcated. A fairly equal development of the 

 branches is further seen in pl. 1, fig. 23, and in pl. 4, figs. 9 — 12. Only rarely is there 

 any considerable difference between the sister-branches. Examples of unequal devel- 

 opment are seen for instance in pl. 1, fig. 22, but even here the narrower branch has more 

 than half the breadth of the thicker one. A conspicuous feature in regard to the mode 

 of branching is that the sister-branches form a very acute angle with each other, being 

 both directed upwards and nearly parallel to each other. Though it is possible that a 

 main stem may sometimes become differentiated in a sympodial manner, the branches 

 thus become lateral retain the characters of the main axis, being upright, simple or 

 sparsely bifurcating and covered with spines. »Several specimens show the circinate 

 curving of the young branch-tops — thus, for instance those in pl. 4, figs. 7 and 8. All 

 the specimens which present this feature and which thus may be regarded as young bran- 

 ches are fairly thick, at least a couple of millimetres in diameter. 



The spines always form a right or nearly right angle to the axis. They are usu- 

 ally straight, linear, gradually tapering from a thicker base. To judge from the shape 

 of the scars, they were rounded in section. Their length varies considerably: from 1 to 



3 mm., 2 mm. being the common average. The shape is also somewhat different in dif- 

 ferent specimens: sometimes the basal part is rather thick in respect to the length, so 

 that the conical shape becomes more pronounced. In other specimens the shape is more 

 narrowly linear. The surface of the spines is faintly striated longitudinally, but there 

 is no evidence of the existence of a vein. 



On the surface of the impressions the spines cannot usually be observed, whereas 

 they appear more or less distinct as seen in profile at their outlines. On some impres- 

 sions of the surface of the stem, however, the position of the spines is shown by markings 

 which may be impressions of the spines themselves or of their thicker bases remaining 

 after their upper parts have been broken off. It is a curious fact that such markings are 

 best seen in specimens which do not, as is usual, show the spines in profile at the contours 

 of the impressions. On impressions of the outer stem-surface, the markings appear as 

 rounded and rather deep hollows. They are shown for instance in pl. 2, fig. 1, and in higher 



