6io Edwards. — Fossil Coniferous Wood from Kerguelen Island. 
‘in every stage from that of charcoal lighting and burning freely when put 
in the fire, to so high a degree of silicification as to scratch glass. A bed of 
shale, several feet in thickness, which was found overlying some of the fossil 
trees, had probably prevented their carbonization when the fluid lava 
poured over them.’ McCormick gives a detailed account (McCormick, 
1842, reprinted in Ross, 1847, p. 74) of the occurrence of wood, and states 
that it was found ‘ both imbedded in the basalt and in the debris below, or 
scattered on the surface amongst the fragments of rocks ’. He refers to the 
shale, but says that no remains of leaves could be discovered in it. (His 
collection contains a few small pieces of sandy shales full of plant impres- 
sions too fragmentary for identification, and labelled ‘ Christmas Harbour ’.) 
Beds of slaty coal were met with at many points under the basalt. Hooker 
(1847, p. 219) writes : ‘ Throughout many of the lava-streams are found 
prostrate trunks of fossil trees of no mean growth, and the incinerated 
remains of recent ones, so that it seems impossible to reckon the period of 
time that must have elapsed between the origin, growth, and destruction of 
the successive forests now buried in one hill.’ 
The fossil wood was also observed by members of later expeditions to 
the island, notably the voyages of the Gazelle (Studer, 1889, p. 61) and the 
Challenger (Murray, 1885, p.349), and the German South Polar Expedition 
of 1901-3 (Philippi, 1908, p. 197). The wood brought back by the Gazelle 
from Christmas Harbour was examined microscopically by Goeppert (1881, 
p. 28) and by Beust (1884, p. 10). The former signalized the presence of 
araucarian wood, but without description or illustration, and, as Gothan 
(1908, p. 13) says, it is doubtful whether Goeppert’s Araucarites Schleinitzii 
et Hookeri refers to one species or to two, and both are really nomina nuda . 
However, Seward (1914, p. 1 1) stated in 1914 that ‘ the examination of 
sections cut from a piece of petrified wood in the British Museum, obtained 
by Sir J. D. Hooker, enables me to confirm Goeppert’s statement as to the 
occurrence of araucarian wood ’, and later these sections were named 
Dadoxylon kerguelense (Seward, 1919, p. 185). Beust in 1884 named some 
of the Christmas Harbour wood Cupressinoxylon antarcticum (Beust, 1884, 
p. 12), but he was only able to examine a few small fragments. He 
described it as having separate bordered pits in one row, abundant resin 
parenchyma, and uniseriate rays 1-8 cells high. The specimens here 
described from the same locality apparently belong to the same species. 
The only other paper on the internal structure of Kerguelen wood 
is by Crie (1889, p. 8), who described as Cupressinoxylon kerguelense some well- 
preserved pieces of silicified wood which had been sent him from London by 
Gardiner and Etheridge. His description is, however, wholly inadequate, 
and his illustrations are poor. He states that C. kerguelense appears to differ 
from all hitherto described species, and then gives a reference to Beust’s 
paper, but does not say wherein the differences lie. According to Crie, the 
