lies Trondhjem 011 ancient, classic soil. South, of this fjord, the
strip of population tapers off along the coast, but in Sogn pushes
far inland. South of the Sogne fjord, the line of habitation
again becomes broader along the coast about Bergen, Haugesund
and Stavanger, and several of the islands outside are quite thickly
inhabited. In the Lister-Mandal division there comes another
broader belt of inhabited country, which, however, diminishes
again in width at Arendal, but then once more becomes broader,
and, with a few off-shoots upwards — large valleys -h merges into
the well-populated districts around and north of the Kristiania
Fjord. This inhabited country continues in an almost unbroken
line through 0sterdalen and Guldalen to Trondelagen, through Gud-
brandsdalen and Raumadalen to the Romsdal division, and through
Yaldres to the country round the head of the Sogne Fjord.
About two thirds of the entire population of the country live
upon the coast and up the fjords, about a fourth part in the
interior lowland districts, while the remainder, about 10 % of the
population, belong to the mountain districts. The dwellings in
the latter extend to a considerable height above the sea, the height
in some places being more than 3000 feet for farms for winter
habitation.
We will now proceed to consider the distribution of the population
from another point of view, namely, its division into country
and town population.
In 1891, the population of the Norwegian towns amounted
to 474,129 persons or 23.70 per cent of the whole population, while
the remaining 76.30 per cent fell to the rural districts., The growth
of the town population in Norway belongs principally to the present
century. In 1801, the towns did not amount to quite 10 °/o
of the population of the kingdom, and not until about 1880 was
20 % exceeded; but at the close of 1896, they amounted to about
26 %. Sweden, and still more Finland, have a relatively less
numerous town population than Norway; Denmark on the contrary,
a very much more numerous one. The average for Europe may
be put at 33 %.- In England, in 1891, between.2/« and 3/i of the
population lived in the towns. Comparison, however, is difficult,
as the idea expressed by the word «town» is very vague. In
Norway, the limit is legal, in several ' other countries only conventional,
every collection of people, that is above a certain size,
being reckoned as a town.
There are now 61 towns' in Norway, as against 42 in the
year 1801. They are nearly all small. In 1891 there were 42
with less than 5000 inhabitants, 9 from 5000 to 10,000, 5 from
10,000 to 20,000, 3 from 20,000 to 50,000, and 2 above 50,000.
The growth of late years has chiefly gone on within the same
groups, whose relative size is thus still almost unaltered. The
three largest towns are Kristiania, which, on the 1st Jan, 1899,
numbered 221,255 inhabitants, Bergen with 68,000 in 1899, and
Trondhjem with 33,033 on the 1st Jan. i897.
The population of Kristiania and Bergen together, amounts
to about half the town population of the country, which, according
to the calculation of the 1st Jan. 1897, amounted to 550,000, but
has since grown a little.
With a few, comparatively slight, exceptions, the Norwegian
towns lie along the coast, the tract from Fredrikshald to Kristian-
sand being thickly studded with large and small towns. The
largest inland towns are the mining town of Kongsberg with about
5500, and Hamar with about 5000 inhabitants. Outside the towns,
the buildings are ^as a' rule scattered, as the Norwegian rural
population does not live, as in several other lands, in villages, but
in solitary farms, with their cultivated land round them. Upon the
coast, however, the fishing population has formed village-like groups
of houses in several places, and these villages have also sprung up
in a few inland places, where industrial undertakings have occar
sioned any considerable concentration, e. g. Lillestrommen and Roros.
III. COMPOSITION OF THE POPULATION.
At the last census, there were enumerated in all 443,317
separate households, 385,220 of which were true family households,
while the corresponding figures in 1876 were respectively
389,611 and 341,806. The average number of persons in each
family household in 1891, was 5.01, reckoned from the domiciled
population, and 5.15 in 1876.
Besides the family households, in 1891 there were 623 other
households (poor-houses, infirmaries, houses of correction, etc.)ji and
57,474 solitary persons, of whom 27,275 were men, and 30,199
women, or respectively 2.82 and 2.92 per cent.of the total population
of each sex. If all the households are taken together, and the
actual population at the time is used as a basis for the calculation,