Jökull - 01.12.1993, Síða 20

trees had grown and hence estimated the latitude and
origin of the wood (Agardh, 1869). The methods de-
scribed in Agardh’s work were later utilised by many
driftwood researchers.
Örtenblad (1881) analyzed over 100 driftwood
samples from the south coast of Greenland, collected
by the geologist Holst in 1880. He applied similar
methods as described by Agardh (1869), but made
more detailed studies. He measured and counted the
tree rings, compared the mean tree-ring widths with
that of trees living in different places in Sweden and
concluded that the origin of the driftwood was north of
66°latitude. He also made a systematic identification
of the samples by using colour differences and analy-
sis of the wood anatomy. Örtenblad calculated the fol-
lowing distribution of species: Larix 44%, Picea 28%,
broad leaf trees 21%, Pinus 4% and Abies 3%. He
concluded that Siberia was the main origin of the drift-
wood on Greenland and that the samples had drifted
from the Ob-Yenisey rivers, north of Svalbard and
southwards to Greenland (Figure 2). The results of
Örtenblad’s studies were later used by Nansen when
planning the FRAM-expedition. The samples used
by Örtenblad were partly re-analysed in Eggertsson
(1994a) using the dendrochronologrcal method.
Lindman (1883) studied driftwood from the coast
of Norway. He also identified fruit and seeds native
to the West Indies and Central America and therefore
concluded that the Larix driftwood he found on the
same coast was a Gulf Stream product from the St.
Lawrence river in Canada. Lindman’s conclusions on
the origin of the driftwood were later criticised by
Ingvarson (1903) who argued that the Larix driftwood
probably originated in the Siberian forests.
Ingvarson (1903 and 1910) made a thorough re-
view of earlier investigations of driftwood and anal-
ysed wood from Kung Karls Land, Svalbard, Jan
Mayen, northeast Greenland, the Yenisey river mouth
at Dickson harbour and from Ellesmere Island. He was
able to identify differences between the wood anatomy
of Larix and Picea, which had been and still remains
a problem in driftwood studies (Bartholin, 1979). He
concluded that the driftwood found on the Arctic is-
lands had not been transported by the Gulf Stream,
but that some, primarily that found on Ellesmere Is-
land, might have originated from rivers draining into
the Arctic seas from the boreal forest areas of North
America (Ingvarson, 1910).
Since Ingvarson (1910) no paper was published
which in detail discussed the identification of drift-
wood species, until Eurola (1971) made a summary of
earlier investigations. This accompanied his own work
on 19 driftwood samples from Agardhbukta on the east
coast of Spitsbergen. Eurola plotted on a map the lo-
cations of species-identified driftwood from previous
literature, plus his own collection. He summarised the
methods used in driftwood studies, made comments
on the origin of the Arctic driftwood and divided the
area north of the Arctic Circle into two, the American
and the Siberian driftwoodregions. Haggblom (1982)
estimated the amount and type of driftwood logs on
raised beaches at different elevations a.s.l. on the island
of Hopen, in the Svalbard archipelago and concluded
that they probably reflected variations in sea ice con-
ditions over time. His idea was based on the fact that
floating wood gradually loses its buoyancy.
Samset (1991) studied driftwood from the north-
west coast of Iceland, dominated by Pinus and Larix
logs, which according to him probably originated in
the boreal forest regions of Siberia.
DENDROCHRONOLOGY AND DRIFTWOOD
Dendrochronology (or tree-ring dating) (e.g. Sch-
weingruber, 1988) is based on the yearly variation of
tree-ring widths. The tree rings are wide or narrow
depending on limiting growth factors. If trees in the
same area show similar tree-ring patterns, the limiting
growth factor is the same over the whole area (e.g.
summer temperature). When this is the case, it is pos-
sible to build up mean tree-ring chronologies that can
be used to date wood material with an unknown age
which originates from the same area.
Giddings (1941) was the first to apply the den-
drochronological method to driftwood. During his
dendrochronological studies in Alaska he noticed that
the tree-ring record at the timberline did not change
rapidly from one locality to another in the Alaskan
interior. Giddings assumed that cross-dating was ap-
plicable to dead logs. He made a collection of samples
from old houses and standing structures built of drift-
18 JÖKULL, No. 43, 1993