A few hours after reading the above article in the October Journal an incident occurred which highlighted the dangers indicated by Captain Kemp. At night, but in clear weather and deep water, and with no navigational hazards, a ship was reported at Green 30, 3 miles, with a C.P.A. of 2 cables to starboard. This was unnecessarily and dangerously close and the officer-of-the-watch (a relatively inexperienced watchkeeper) proposed standing on to 1 mile and then altering 10 degrees to port to open the range.

There are occasions when it can be useful to plot bearings from objects not shown on the chart, for example a radiobeacon or a lighthouse outside the coverage of the chart in use. The traverse table may be used for this purpose in the following manner:

1. Establish the latitude and longitude of the object (radiobeacon, lighthouse, &c).

2. Choose a convenient latitude from which to plot the bearing and calculate the difference of latitude between it and the object.

3. With the bearing as course extract from the tables the departure corresponding to this d. lat.

4. For the middle (or mean) latitude derive the corresponding d. long. and apply it to the longitude of the bearing object.

5. Plot the bearing, as a position line, through the chosen latitude and the longitude thus derived.

As the hitherto anonymous reviser who prepared the current G.H.A. edition of Martelli's Tables, was interested, instructed and surprised by Captain Cotter's article in the Forum on this subject. I was surprised that he should write of them as if they were things of the past. In fact they are available off the shelf at my local chart agents, and no doubt elsewhere, so a demand presumably exists today, and they are in current use. From some of the comments which he makes, and the extract from Table 5 which he reproduces, it appears that he has not seen the current edition.

Much of Captain Wylie's note (this Journal, 27, 111) is undoubtedly correct, particularly the part headed ‘automatic versus manual’, which restates the strong operational requirement for automatic plotting. The earlier part, however, contains some truth, some error, and shows a notable omission. A detailed analysis would take far too much space, so I will deal with three points only.