I 



ON GKAPIIIC METHODS IN MECHANICAL SCIENCE. 



383 



the lines will be closer together, and the diagram will tend to lose its 

 clearness. These results can, however, be transferred to a model, each 

 difference of level being represented to scale, and the different heights 

 heing shown as a series of steps. In this case, however, not the slightest 

 confusion arises in producing every possible variation of height, which 

 was done in one section of a model exhibited before the Section by 

 forming a continuous slope by means of plaster. Measurements can be 

 taken at any point of any of the three variable quantities, namely, the 

 day of the month, or hour of the day, or height of the tide, when any two 

 out of the three are given. The model in question shows that the diagram 



Fig. 11. — Tide Diagram, Liverpool, November 1888. 



NIGHT 



NOON 



NIOHT 



/ Z 5 S J e 1 S 3 ro S> IZ 13 !4- li 16 77 S JS Zj) 2J ZZ Z3 Z4- S Z€ Z7 ZS Z9 30 



The curves give differences of 4 feet from Old Dock-sill. Lines marked give datum 



level. Lines marked —. — • — • give levels below datum level. Thick lines give levels 

 above datum level. 



Note. — The Old Dock-sill is 4-65 feet below ordnance datum. 



may be practically regarded as an example of the use of contour lines ; and 

 just as contour lines are generally employed for giving various heights of 

 mountains, and indicate the slope or steepness of various parts by their 

 proximity to each other, so the rapidity with which the tide rises and falls 

 on certain days of the month is equally well shown by the closeness of the 

 lines of the tide diagram. 



For instance, the tide rises and falls much more rapidly on the 4th 

 than on the 11th, for in the former case the intersections of the curves 

 by the ordinate are much more close together than in the latter. 



This use of contour lines is generally employed on maps, not only 

 for representing the heipjhts of mountains, but for showing the density of 



