KEW OBSERVATORY AND METEOROLOGY 235
compendious yearly volume. I f so, the tracings would require very much more reduction in breadth than in height, for the photographic mark made by the recorder was so broad that the scale of the tracing had to be proportionately wide open ; otherwise the neighbouring irregularities would blur together. A sharp line drawn along the middle of the tracings might, however, be much compressed laterally and yet show all the irregularities distinctly. I designed a compound drill pantagraph for the purpose, which reduced the tracings in height independently of the reduction in length. One part of the machine worked the drill forward and backwards, the other part moved the plate from side to side upon which it worked. The result was to express the tracings by fine grooves cut into a piece of soft metal. These were again reduced by an ordinary pantagraph. The whole process required thinking out in numerous details, but it proved quite a success. It is described in the annual Report of the Meteorological Office for 1869.
Squares of zinc, one for each day, were grooved by the drill pantagraph so as to show every one of the data without confusion. They referred to Wind Velocity and Direction, Barometric Height, Rainfall, Dry and Wet Thermometer, together with a line to show the amount of Humidity in the air, which was mechanically calculated from the combined traces of the two thermometers. These squares were placed beneath a large and beautifully designed German pantagraph, whose pointer was directed along the grooves in the zinc, while the diamond point of the scribe scratched the varnish on a copper plate, which