On the fourth day we lay quiet all day long, with a slight breeze from Princess Marie Bay setting us slowly eastward; but, as the sun was shining, we utilized the time in drying our clothing, wet and soggy from the almost continuous rain and snow of the previous two days. As it was still summertime in the Arctic, we did not suffer from cold. The pools between the ice floes were slowly enlarging, and at nine in the evening we were on our way again, but at eleven we ran into a thick fog. All night we bored and twisted through the ice, which, though thick, was not heavy for the Roosevelt, and only once or twice we had to back her. An ordinary ship could have made no headway whatever.

Wardwell, the chief engineer, stood his eight-hour or twelve-hour watch the same as his assistants, and during the passage of these dangerous channels he was nearly always in the engine-room, watching the machinery to see that no part of it got out of order at a crucial moment—which would have meant the loss of the ship. When we were between two big floes, forcing our way through, I would call down the tube leading from the bridge to the engine-room:

"Chief, you've got to keep her moving until I give you word, no matter what happens."

Sometimes the ship would get stuck between the corners of two floes which were slowly coming together. At such a time a minute is an eternity. I would call down the tube to Wardwell, "You've got to jump her now, the length of fifty yards," or whatever it might be. And I could feel the ship shaking under me as she seemed to take the flying leap, under the impulse of live steam poured directly from the boilers into the fifty-two-inch low-pressure cylinder.

The engines of the Roosevelt have what is called a by-pass, by which the live steam can be turned into the big cylinder, more than doubling the power of the engines for a few minutes. This simple bit of mechanism has saved us from being crushed flat by the ice on more than one occasion.

The destruction of a ship between two ice floes is not sudden, like her destruction by a submarine mine, for instance. It is a slow and gradually increasing pressure from both sides, sometimes till the ice meets in the vitals of the ship. A vessel might stay thus, suspended between two floes, for twenty-four hours—or until the movement of the tides relaxed the pressure, when she would sink. The ice might open at first just sufficiently to let the hull go down, and the ends of the yards might catch on the ice and break, with the weight of the water-filled hull, as was the case with the ill-fated Jeannette. One ship, in the Gulf of St. Lawrence, was caught in the ice and dragged over the rocks like a nutmeg over a nutmeg grater. The bottom was sliced off as one would slice a cucumber with a knife, so that the iron blubber tanks in the hold dropped out of her. The ship became nothing but the sides and ends of a box. She remained some twenty-four hours, gripped between the floes, and then went down.

On the 22d of August, the fifth day, our lucky stars must have been working overtime; for we made a phenomenal run—more than a hundred miles, right up the middle of Kennedy Channel, uninterrupted by ice or fog! At midnight the sun burst gloriously through the clouds, just over Cape Lieber. It seemed a happy omen.

Could such good fortune continue? Though my hopes were high, the experience of former journeys reminded me that the brightest coin has always a reverse side. In a day we had run the whole length of Kennedy Channel, and immediately before us there was only scattered ice. But beyond lay Robeson Channel, only some thirty miles away, and the navigator who knows Robeson Channel will never be sanguine that it has anything good in store for him.

Soon we encountered both ice and fog, and, while working slowly along in search of an opening, we were forced clear across to the Greenland coast at Thank God Harbor, the winter quarters of the Polaris in 1871-72. I have mentioned the lane of water which often lies at ebb tide between the land and the moving central pack; but the reader must not fancy that this is an unobstructed lane. On the contrary, its passage means constant butting of the smaller ice, and constant dodging of larger pieces.

Of course the steam is up at all times, ready, like ourselves, for anything at a moment's notice. When the ice is not so heavy as to be utterly impenetrable, the ship under full steam moves back and forth continually, butting and charging the floes. Sometimes a charge will send the ship forward half her length, sometimes her whole length—sometimes not an inch. When, with all the steam of the boilers, we can make no headway whatever, we wait for the ice to loosen up, and economize our coal. We do not mind using the ship as a battering ram—that is what she was made for; but beyond Etah coal is precious, and every ounce of it must yield its full return of northward steaming. The coal at present in our bunkers was all that we should have until our return the following year, when the Peary Arctic Club would send a ship to meet us at Etah.