Holden. — Some Fossil Plants from Eastern Canada. 247 
Tylodendron with the Araucarians. Several varieties of leafy branches, 
known as Walchia , and definitely associated with Tylodendron pith casts, 
have been described, all bearing a close resemblance to different species of 
Araucaria. Of their fructifications little is known, further than that, as 
shown by Zeiller, 1 the scales of the female cone bear single seeds, another 
Araucarian feature. If all these criteria are reliable, the presence of Tylo- 
dendron in Permian strata bears out the orthodox view that the Araucarineae 
are the oldest living family of the Coniferales. 
When this evidence is weighed more carefully it is not so convincing. 
The worthlessness of Felix’s distinction, based on whether or not the radial 
wall is covered with pits, Gothan has already detected. He points out 2 that 
Dadoxylon stephanense and D. subrhodeanum are described by Grand’Eury 3 
as encircling Artisian pith casts ; and yet the pitting of the radial walls of 
the tracheides is usually uniserial. As regards the shape of the mouth of 
the pit, Gothan may be correct in stating that the pits of Cordaites always 
have elliptical openings, yet in the Conifers there seems to be no invariable 
rule. For example, Agathis has sometimes elliptical and sometimes circular 
openings, and such is the case throughout the Abietineae and Cupressineae. 
The other lines of evidence adduced above demonstrate that the relation- 
ship between Tylodendron and the Cordaitales cannot be close, but do not 
prove a closer between Tylodendron and the Araucarians. All living Conifers 
have low, uniseriate rays, instanding protoxylem strands, and no extensive 
development of primary wood. Though the character of the pitting of the 
tracheides suggests the Araucarian Conifers, recent investigations of Mesozoic 
woods have shown that closely compressed and alternating pitting is not 
the primitive condition for the Araucarineae. 4 Further, there are woods of 
the Tylodendron type extending as far back as the Culm, yet no advocate 
of the antiquity of the Araucarian line would suggest that it extends as far 
as that. We can do no better than agree with Gothan 6 when he writes 
that ‘ Walchia has not been proved to belong to the Araucarineae, and that 
the opposite is much more probable’. 
Summary. 
1 . There are abundant remains of Tylodendron in the strata of the 
south shore of Prince Edward Island. 
2. Since Tylodendron is characteristic of the Permian, there can be no 
question that these strata are of that age. 
1 Zeiller : Etudes sur la flore fossile des depots houillers et permiens des environs de Brive. 
Paris, 1892. 2 Gothan : loc. cit., p. 15. 
3 Grand’Eury: Flore carbonifere du Dep. de la Loire, p. 257. 
4 Jeffrey, E. C. : The Araucarioxylon Type. Proceedings Am. Academy of Arts and Sciences, 
vol. xlviii, No. 13, Nov., 1912. 
2 Gothan, Walter: loc. cit., p. 13. 
