SEA LOG AND SOUNDING MACHINE. £47 



wake of the vessel, it had a length of fifty yards of cord, or 

 jnore, to carry round with it every time it revolved. The 

 friction caused by this operation was such as to preclude all 

 hopes of accuracy ; it was useless in an agitated sea, the 

 rope was very liable to kink, and in fast sailing the rotator 

 would sometimes fly out of the water*. These circumstan- 

 ces rendered it impossible, that the rotator should make the 

 same number of revolutions in passing through a given 

 space, under diff'erent velocities ; and hence inaccuracy was 

 unavoidable. To get rid of this friction of the long line, 

 the rotator has, in some instances, been enclosed in a cylin- 

 der, and a register been attached to the outside. But though 

 the defect of excessive friction was thus surmounted, still 

 greater inconveniences resulted. It may be sufficient to 

 mention, that the cylinder, not presenting itself horizontally 

 in the water, was liable to alter its position whenever the ve- " 



locity of the vessel was changed, which caused an eddy, or 

 dead water, to remain in the cylinder; and; of course, the 

 rotator was liable to err, in proportion as the cylinder lost 

 its horizontal position. 



After thus hinting at the imperfections of other previous 

 methods of constructing logs, it remains to point out 

 wherein Masscy's plan differs. 



Friction is the principal cause of mechanical theories va- Difference of 

 rying so widely from actual experiment. In some machines | " present 

 one third is allowed for its effect, while the operation in 

 others is nearly suspended, and what appeared very plausi- 

 ble in theory, is found totally useless in practice. Thus the Friction trifling 

 friction on a rope long enough to extend beyond the eddy of 

 the vessel's wake would, in many circumstances, on the old 

 plan, totally impede the action of the rotatorf. Under this 



impression, 



• • Smeaton, in the account of hiS experiments on Sauniarez's log, in 

 the Philosophical Transactions, observes on this subject " Upon mak- 

 ** ing up the account of this run, 1 found the number of rotations were 

 <* less by one full third than they ought to have been, compared with 

 ** the former observations, which afford6d me a convincing proof, that 

 <* this instrument was considerably retarded in quick motions " 



•f Smeaton, in the account of his expieriments in the work before 

 quoted observes : " During this run, 1 observed that the resistance of 

 *• the water to the line and plate was very considerable, and increased the 



** friction 



