The swamp occasioned many anxious nights, and much burning of midnight oil. At places it appeared to be bottomless. The ballast locomotive would haul train after train-load of spoil excavated from the ballast-pit, and push it cautiously along to the end of the dump, where the trucks would be discharged. The rubble would rush down the declivity, and as it came into contact with the surface of the morass there would be a wicked squelch. Then the bog would open, and slowly, but surely and silently, the discharged mass would disappear into the viscous mass until the last vestige had slipped from sight, and the slime had rolled over the spot, concealing all evidences of the few hundred tons of material emptied on to the spot but a few minutes before. The engineer would sound the bog anxiously for signs of the bottom. Yes, he could feel it all right--10, 15, perhaps 20 feet below the surface. The trains would continue to rattle up and down with heavily laden trucks, and send the contents crashing pell-mell into the swamp below. Ten train-loads of gravel, rock, and what not would disappear from sight, and the engineer would probe the treacherous sponge once more. But the soundings did not vary a foot. Where had the dump gone? The ballast had sunk simply to the bottom of the bog, and had spread itself out on all sides, finding its own level like water. The bed of the morass was as broken as the hill-side near by, and was intersected in all directions by ruts and gullies. Until these holes were filled, there could be no possible hope of the embankment appearing above the surface of the bog.... While the upper stretches of New Ontario and Quebec were occasioning the engineers many anxious moments, owing to the eccentricities of the muskeg and swamp, the graders advancing eastwards from Winnipeg were in close grips with rock, which offered a most stubborn resistance.... For hour after hour, day after day, month after month, nothing was heard but the chink-chink of drills and the devastating roar of explosive with its splitting and disintegrating work. Advance was exceedingly slow, some of the blasts requiring as much as six weeks or more to prepare, and then only breaking up sufficient of the granite mass to permit of an advance of about 200 feet.... It was on work of this nature where the greatest number of accidents occurred, the majority of which might have been avoided had the men engaged in the operations displayed but ordinary care.... Dynamite was responsible for more deaths on this undertaking than any other accident--the collapse of the Quebec Bridge notwithstanding--and sickness combined....