110 
This species was recorded by Dawson 1 from Mackenzie river under 
the names of Populus arctica and Populus hookeri, and it has also been 
recorded from many localities in Alaska, Greenland, and western Canada. 
The leaves listed as Grewia crenata in my preliminary list of the fossil 
plants from Newhykulston creek, and the occurrences listed by Penhallow 
and referred to Grewia crenata , namely, Horsefly river and Porcupine 
creek, also probably represent Trochodendroides. 
The family Trochodendraceae consists of 7 or 8 species of temperate 
central and eastern Asia segregated in the genera Tetracentron, Cercido- 
phyllum, Euptelea, and Trochodendron. They show primitive features 
in the absence of vessels in the wood, a unique feature among angiosperms, 
and in having the growing point enclosed in the petiole of the terminal 
leaf. This last feature as well as the leaf form are present as early as the 
late Lower Cretaceous, where they were again referred to Populus or 
Populophyllum, and it seems probable that the existing family represents 
the restricted progeny of this once Holarctic type. 
The present species is present in the collections studied by me from 
Kitsilano and Newhykulston creek. 
Family, magnoliaceae 
Magnolia nordenskioldi Heer (?) 
Magnolia nordenskioldi Heer, FI. Fdss. Arct., Bd. 4, abt. 1, p. 82, PI. 21, 
fig. 3, PL 30, fig. 1, 1877; Bd. 5, p. 46, PI. 3, fig. 2b, 1878; Bd. 6, 
abt. 1, p. 13, PI. 5, fig. 10, 1880; Bd. 7, p. 123, PI. 82, fig. 1, PI. 108, 
figs. 2, 3, 1883; Dawson, Trans. Roy. Soc., Can., vol. 1, p. 33, 1883; 
Lesquereux, U.S. Nat. Mus., vol. 5, p. 448, PL 10, figs. 7-9, 1883; 
Penhallow, Rept. Tert. PL Brit. Col., p. 62, 1908. 
This species was based upon numerous fragmentary specimens of a 
variable leaf, which has been recorded from the Eocene of Alaska, the 
Mackenzie basin, Spitzbergen, Sakhalin island, and Greenland. There 
are two specimens in the collection from Kitsilano that appear to repre- 
sent this species, but they are poorly preserved and not certainly identified. 
Liriodendron (?) sp. 
There are two incomplete specimens in the collection from Joseph 
creek showing a long, stout petiole and the basal part of two leaves, which, 
as far as they go, are indistinguishable from those of the genus Lirioden- 
dron. In the absence of more representative material I hesitate to do 
more than call attention to them. 
There is a considerable gap in the most interesting geological history 
of this genus in North America following its great abundance during the 
Upper Cretaceous, and there is no reason why it should not be present in 
our more northern Eocene as it is found at that time in Greenland. Lirio- 
dendron is another genus, like Sassafras, once abundant throughout the 
Holarctic region and now restricted to southeastern North America and 
southeastern Asia. 
1 Dawson, J. W., Trans. Boy. Soc., Can., vol. 7, seo. 4, p. 71, PI. 10, figs. 2-5, 1889. 
