Log, and Log Line, an apparatus used in connection with the half-minute glass for obtaining the approximate rate of movement of a vessel through the water. The log is a triangular or quadrangular piece of board, one side of which has a circular edge, and is weighted with lead, so as to cause the piece to sit upright when thrown into the water. It is attached by cords from its corners to the log line, which is a stout cord about 150 fathoms long, divided by knots or slips of leather into spaces called knots, and wound on a reel which revolves with freedom. Its use is called "heaving the log," and consists in dropping the wood over the stern of the vessel, with a quantity of the line sufficient to reach from the vessel to the log, at the instant the half-minute glass is turned up. The reel is held up so that the line may run off freely as the vessel moves away from the log; and as the last sands run through the glass, the reel is instantly stopped. The number of knots run off in the half minute indicates the rate of motion of the vessel. This method of measurement is very inaccurate, a heavy sea sometimes throwing the log after the ship, while a head sea may carry it in the opposite direction.

The glass also measures the half minute differently in damp and dry weather, and the line is liable to change its length. Various empirical allowances are made, which add but little to the correctness of the apparatus. It is not known when or by whom this contrivance was invented. Humboldt says that in all writings on the subject, including the " Encyclopaedia Britannica," he found the erroneous opinion expressed that the log was not introduced before the end of the 16th or the beginning of the 17th century, while it is certain that Pigafetta, the companion of Magalhaens, early in the 16th century, speaks of the log (la catena a popa) as of a well known means of measuring the course passed over. Purchas makes mention of it in 1607; but the length of a degree of the meridian not being then determined, its divisions were necessarily inaccurate. They were corrected in 1635 by Norwood. The length of a sea mile is now estimated at about 6,086 ft.; and as the length of the knot is intended to bear the same proportion to this that half a minute bears to an hour, the measurement of the knot is properly 51 ft. Each one is divided into 10 parts called fathoms. For glasses which run out in 28 seconds, the length of the knot should be 47.6 ft. - Numerous substitutes for the log have been contrived.

The best of these is that of Massey. A box shaped like a wedge is provided with a spindle to which four wings are fixed spirally. With this are connected registering wheels somewhat on the plan of those of the gas meter, their object being to record the number of revolutions of the spindle. This is carried round by the motion against the water as the box is towed astern by a stout line 60 fathoms long. The box is hauled in, and the record noted whenever the course is changed; but while the ship runs full three knots the register is not reset except once every 24 hours. At a less rate its indications are uncertain from not towing horizontally.