Phytography of the Arctic American Archipelago 
23 
The Arctic American Archipelago, such as we now know it through the work 
carried out by about fifty different expeditions, aniong which tlie niore important 
are mentioned above, forms a wide complex a islands of very different size. Its 
southernmost point is Hatton Headland (Cape Best) on Resolution Island in about 
61 Va" N., its most northern are Cape Columbia, H3*^8', in Ellesmereland and the 
north shore of Ward Hunt Island, lying off the coast of the same land. The most 
extreme eastern points are Cape Walsingham in Baflin Land and Cape Union m 
Ellesmereland, both almost exactly in 61" W. fr. Greenwich. Farthest- to the west 
reaches Cape Kellett, about 1250W., in Banks Land. The joint area of all the 
islands wliich is indeed not even approximately known, may be between 600,000 
and 700,000 square miles. As new lands have been discovered so lately as about 
ten years ago, by the last expedition which went up to the northwestern islands, it 
might seem possible tliat more islands could still be found in that direction, and 
G. Braun, indeed, has lately in a little populär book (Erforscli. d. Pole, p. 11) 
drawn the conclusion, that the discoveries of Svkrdrup and Isachsen make pro- 
bable »das Auftreten weiterer Landmassen im Parry-Archipel». In arriving at this 
conclusion he has, however, overlooked a wery important fact, viz., the ice-condi- 
tions on these northwestern shores. As w^ell on the west coasts of the Ringnes 
Islands and of Heiberg Land as also at the northern end of Nansen Sound tlie ice 
was old and heavv, very uulike that of the sounds and basins between the islands. 
If I8ACHSEN'8 aud SvEKDRUp'8 dcscriptious of the ice are compared with those of 
Aldrich, M'Clure, M'Clintock, and others from the north shore of Ellesmereland 
and from Banks Land and Prince Patrick Island, they wiU in my opinion, support 
the assumption, probable in itself, that tlie tolerably continuous borderline of the 
Archipelago from Cape Bathurst up to the north coast of Ellesmereland (and Green- 
land) also forms the extreme limit of the whole American landcomplex. Here the 
heavy pack of the Polar Basin sets in against the Continental shelf, which will, 
most probably, some time be found to run parallel to the trend of the coast. 
The wide almost triangulär expause of the Archipelago is very irregularly 
separated by sounds and channels of different breadth into islands of varying size 
and outline. The largest is Baffin Land, about 150,000 square miles and including 
nearlv a fourth of the whole area. Next come Ellesmereland and Victoria Land, 
each about 60,000 square miles, and ten more are of a rather considerable size. 
As a look on the accompanying map shows, the Archipelago may naturally be 
divided in five or six parts or separate groups. Three of these lie south of a well- 
defined boundary formed by broad channels, beginning with M'Clure Channel in 
the west and ending in Lancaster Bound in the east. Among these southern divi- 
sions the westernmost is formed by Banks and Victoria Lands and limited to the 
east by Victoria Strait and M'Clintock Channel. Next comes a section including 
Prince of Wales Island and North Somerset, to which King William Land may also 
l)e reckoned. Then, however, Boothia Felix ought also to form a part of this sec- 
tion, for nothwithstanding its connection with the mainland it seems, as far as its 
