56 Life and Letters of Francis Galton
not afford the results that seamen principally require; they are only data from which those results might be calculated by some hitherto unexplained process, which, we can easily foresee, must be an exceedingly tedious one."
It is this process which Galton proceeds to unfold for moderate winds in the case of a "merchantman of the class that usually navigates the Atlantic." To carry it out we require to know: (1) the proportionate time (or the relative frequency) that the wind blows in a given area from each of say eight points of the compass, (2) the number of miles that the particular ship will make in an hour at each angle to the wind. Combining these two results we can measure for the average of the winds in that area the average progress of the ship towards each point of the compass in an hour. If the distance reached' in an hour be plotted from a centre in the arc, we obtain a closed curve whose radius vector measures the efficiency of the ship in that particular area for a particular course. If now the chart be divided up into areas and in each area be placed the corresponding polar diagram, we have converted a wind chart diagram into a passage chart diagram. A navigator now plots his proposed course across these areas, and sets off with his compasses the distance run per hour in the direction of the course from the nearest polar diagram. In this way he is able to calculate the average time on the proposed course and can compare it with the time on other courses.
"He will thus be able to select the quickest out of any number of routes that may be suggested •to him, and to determine, on the most trustworthy of existing data, what is the best course to adopt in, sailing from one part of the ocean to another."
Galton suggests the modification of the polar diagram when (a) force of wind and (b) current are taken into account.
The next paper on this subject was published in the Minutes of the Meteorological Council for December 2, 18722. In this communication Galton advances a considerable stage further. The Meteorological Office had sorted out the whole of the data for direction and force of wind and for current into "single degree squares." Thus the resultant direction and strength of current, the average force of the wind and its proportional directions were more or less accurately known for each area, for each month of the year. Galton now terms the polar diagrams of his earlier paper "isodic curves" or briefly "isods." He calculates them for the month of January for "2° squares" from Longitude 0° to 10° N. and from Latitude 20° to 30° W., allowing for current, and-force of wind as well as direction, and taking as his standard type the " Beaufort ship." The rays now represent the average space run in 8 hours, and Galton enters into details of how to construct 'isodic' charts and passages. He seems, however, to have been in some doubt as to whether his name 'isod' was appropriate. In his own copies of this paper, he questions in pencil whether the word should not be 'ishodic.' But another doubt must have arisen in his mind; his isods did not represent equal paths, but the paths
° Not the distance traversed, because to reach a given point the ship will generally have to tack.
2- Presented to Sir Edward Sabine, the Chairman, Mr Galton, Major-General Smythe and Sir Charles Wheatstone.