element
diminished. On the right bank of the river grew magnificent specimens of
the ulmaceae tribe, the precious elm, so valuable to builders, and which
withstands well the action of water. Then there were numerous groups
belonging to the same family, among others one in particular, the fruit
of which produces a very useful oil. Further on, Herbert remarked the
lardizabala, a twining shrub which, when bruised in water, furnishes
excellent cordage; and two or three ebony trees of a beautiful black,
crossed with capricious veins.

From time to time, in certain places where the landing was easy, the
canoe was stopped, when Gideon Spilett, Herbert, and Pencroft, their
guns in their hands, and preceded by Top, jumped on shore. Without
expecting game, some useful plant might be met with, and the young
naturalist was delighted with discovering a sort of wild spinach,
belonging to the order of chenopodiaceae, and numerous specimens of
cruciferae, belonging to the cabbage tribe, which it would certainly be
possible to cultivate by transplanting. There were cresses, horseradish,
turnips, and lastly, little branching hairy stalks, scarcely more than
three feet high, which produced brownish grains.

"Do you know what this plant is?" asked Herbert of the sailor.

"Tobacco!" cried Pencroft, who evidently had never seen his favorite
plant except in the bowl of his pipe.

"No, Pencroft," replied Herbert; "this is not tobacco, it is mustard."

"Mustard be hanged!" returned the sailor; "but if by chance you happen
to come across a tobacco-plant, my boy, pray don't scorn that!"

"We shall find it some day!" said Gideon Spilett.

"Well!" exclaimed Pencroft, "when that day comes, I do not know what
more will be wanting in our island!"

These different plants, which had been carefully rooted up, were carried
to the canoe, where Cyrus Harding had remained buried in thought.

The reporter, Herbert, and Pencroft in this manner frequently
disembarked, sometimes on the right bank, sometimes on the left bank of
the Mercy.

The latter was less abrupt, but the former more wooded. The engineer
ascertained by consulting his pocket-compass that the direction of the
river from the first turn was obviously southwest and northeast, and
nearly straight for a length of about three miles. But it was to be
supposed that this direction changed beyond that point, and that the
Mercy continued to the north-west, towards the spurs of Mount Franklin,
a