PAEKA DECIPIENS. 11 



represented by their short stumps, are given off at intervals of about 11 cm. Apparently 

 they are alternate and opposite in arrangement, but this may be the result of displace- 

 ment. The iinion of these branches with the main axis is not abrupt — i.e., at right 

 angles — but the separation is gradual, as v\^e find in the lateral members of many aquatic 

 plants having horizontal stems, thus suggesting possible similarity of organisms. And if 

 this idea were extended, it would be quite possible to show a somewhat well defined 

 general resemblance to such horizontal stems as are found in Marsilia and Pilularia. 



No leaves in silii have been found, and it is therefore impossible to decide, except so 

 far as the very limited value of association may afford testimony, as to the character of 

 the foliage produced by this stem. A careful search has also failed to disclose any attach- 

 ments having the nature of inflorescence or fruit. 



That the stem was not highly vascular would seem to be indicated by the extreme 

 compression which it has suffered, and, so far as it goes in connection with what has 

 already been stated, this might also be taken as evidence of the possibly aquatic charac- 

 ter of the plant. In a recent communication from Mr. Eeid he expresses a similar view. 



Fragments of Leaves, Nos. 3 and 4. — Associated with the large stems arc portions 

 of linear leaves. They are generally about 4 to 6 cm. long, and have a uniform width of 

 1 cm. They show one end broken off, and the other somewhat rounded. The carbon- 

 aceous substance is sometimes conspicuous, sometimes replaced by red oxide of iron. No 

 superficial markings are to be seen. These remains are referred to in the present paper by 

 Sir William Dawson as Cordailes ausçuslifolia, from their general resemblance to the leaves 

 of that plant, but without any intention of establishing identity between the two. There 

 is no direct evidence to show what these leaves were derived from. 



The greater number of the specimens examined show impressions or carbonaceous 

 remains of long, linear bodies, either slender stems or narrow leaves. They vary in width 

 from 1 to 3 mm., but show a general tendency towards an average width of 2 mm. As a 

 rule no structure is apparent, though in a few cases fine parallel markings have been 

 observed, suggesting similarity to the nerves of a linear leaf No terminations of these 

 organs have been found, though in the smaller stem — No. 2 — the two lateral members 

 appear to be the same as these organs. It would, therefore, seem highly probable that 

 these remains represent linear leaves belonging to horizontal stems of the dimensions and 

 character of the smaller stems already described. 



The larger leaves (Cordaites) cannot be definitely connected with any stem, but the 

 constancy of their association with the large rugose stems, with which the narrower 

 leaves do not appear to be found, and their close similarity to the latter, would seem to 

 suggest that they are in reality the foliage of those stems. In this connection it may also 

 be well to point out the fact that in his visits to the quarries where Parka is found, Mr. 

 G-raham "got bits of riband-leaves, narrow stalks and associated Parka, the latter in com- 

 parative abundance, and nothing else." 



As pointed out by Sir "William Dawson, remains of Psilophylon are very numerous in 

 connection with Parka, being represented both by stems and fragments of leaves. It is, 

 therefore, quite possible that some of the narrow leaves referred to above as belonging to 

 Parka may in reality belong to Psilophyton, though I think it hardly probable. On the 

 other hand, one specimen from Caithness, Scotland, received from Mr. Reid, shows a tuft 



