12 UNITED STATES COAST AND GEODETIC SURVEY. 



ited in 1908 at the Franco-British Exhibition in London, where it 

 was awarded the grand prix. It is called by him the Universal tide 

 predictor. The principles of its construction, as will be seen from 

 the illustration (plate 4), "are the same as in the India office tide 

 predictor." It provides for 40 components, '^33 of which are actu- 

 ally geared up and vacant places for the gearing of the remaining 

 seven components have been left for the insertion of compound 

 tides which may be found sufficiently large to warrant their inclusion 

 in the predictions." The components now represented are shown 

 in Table B. 



AU these machines sum mechanically by means of a wire or chain 

 fixed at one end and laid under and over the puUeys representing 

 the speeds, amplitudes, and epochs of the components, the weU 

 known series A cos (at +a) +B cos (ht + ^) + C cos (ct + y) + . . ., 

 the free end of the chain, provided with a suitable pen, moving verti- 

 cally and thus tracing the tide curve upon a strip of paper moving 

 laterally at uniform speed. This tide curve must be read or scaled 

 off to obtain the heights of the tides. The times of the high and low 

 waters may also be obtained approximately from the sheet by 

 applying the rate of motion of the paper under the curve pen; for 

 the exact times the machines require a second setting and tracing 

 of a curve representing the first derivative of the above series. 



THE FERREL OR UNITED STATES COAST AND GEODETIC SURVEY TIDE- 

 PREDICTING MACHINE NO. 1. 



To measure these curves and prepare the results for the printer 

 requires far more time than was consumed in their production. This 

 led Prof. WiUiam Ferrel of the United States Coast and Geodetic 

 Survey, when the needs of the Survey required a tide-predicting 

 machine, to plan an apparatus which, instead of tracing a curve at 

 aU, would indicate to the operator for copying directly upon the 

 printer's form the numerical data required for the tide tables. The 

 contract for its construction was given to Fauth & Co., of Washing- 

 ton, D. C. It was completed in 1882 and used for preparing the tide 

 tables published by the Survey from 1883 to 1910. A description 

 with illustrations is given in Coast and Geodetic Survey Report, 1883, 

 pages 253-272. This machine, called by Prof. Ferrel the maxima 

 and minima tide-predicting machine (see plate 5), differs materiaUy 

 from those made in England. Its components, instead of moving at 

 their respective speed ratios, move at speeds representing the differ- 

 ences between them and that of the principal lunar component M^, 

 the motion of the latter being represented by a pointer on the face 

 of the machine. The components represented are given in Table B. 

 The most noteworthy innovation introduced in this machine was the 

 addition to the set of cranks summing the cosine series, as used in 



