50 



ANIMAL MECHANICS. 



The experiment detailed in the note,* made under the 

 direction of Mr. Maclaren, was communicated to me after I 

 had made the preceding calculation, and seems to confirm the 

 result arrived at as to the resistance overcome in this kind of 

 labour. 



* Mr. Maelaren's Experiment. — " An eight-oared racing boat, weighted with 

 sandbags to represent an u st. 3 lb. crew (the weight for which she was built), 

 and steered by an 8 st. coxwain, was towed over part of the Oxford course, where 

 the water is straight, broad, and deep. 



" The four-oared boat by which she was towed was itself towed by men on 

 the bank, and kept in a straight course by a coxwain. The eight-oar was kept 

 as nearly as possible in a line with the four-oar by the coxwain placed on board 

 for that purpose. 



" The towline from the four to the eight was fastened to the bow-oar's thwart 

 in the eight, exactly on a line with the keel, and the strain measured by a dy- 

 namometer (a Salter's spring-balance), interposed between the end of the tow line 

 and the four-oar. 



The distance travelled, 560 yards. 



The time occupied, 6 m . 20 8 . 



The average strain on the dynamometer, . . 7 lbs. 



" There was a light side wind, but not enough to ruffle the water, or seriously 



interfere with the experiment. The strain was measured when the keels of the 



two boats were as nearly as possible in the same straight line. The course was 



against the stream, which is very slight." 



In this case, the velocity expressed in knots per hour is 



560 x 60 x 3 , . . , 



— -= 2.65 knots per hour. 



19 x 2000 



In the case already considered, the velocity was assumed to be one knot in 

 seven minutes, or 



y = 8.57 knots per hour. 



Hence it follows (assuming the resistance to increase as the square of the ve- 

 locity) that the resistance at one knot per seven minutes would be 



1ST- 



7 lbs. = 73.21 lbs. 



Professor Rankine, in a letter addressed to me, 12th September, 1866, sug- 

 gests that I ought to have reduced Weisbach's coefficient from 0.0036 to 0.0030, 



