TIME-KEEPING IN GREECE AND ROME. 395 



iDclinod to bolicve, in one of the. most remarkable instances of retarda- 

 tion of invention that history records. It was not nntil Enrope had 

 emancipated herself from slavery to this most awkward of time systems 

 that modern time-keeping; became possible. For many centniies in- 

 vention was as it were thrown off the scent by the necessity of con- 

 verting the regular and nniform motions which could be given to 

 mechanism into means for displaying the ever- varying hours of the Ro- 

 man system. 



The word "liora," proposed by Hipparchus to express these divisions 

 of the day, was adopted in its new sense by Greeks and Romans simul- 

 taneously and has ev^er since held its place in all the languages of 

 Europe. In fact it was used in two senses ; in its signiticance of the va- 

 rying Roman hour it could not be employed to define exact intervals of 

 time; when employed for that purpose it expressed exactly what we ex- 

 press by it uow, — the twenty-fourth i)art of a civil day. The passage in 

 Pliny I have quoted is not intelligible unless the word " hour" is em- 

 ployed in this sense. 



Enough was said in the early part of this paper to show the line in 

 which the clepsydra developed, the water-clock at Canton and that in 

 the Tower of the Winds at Athens being examples of it in a fairly per- 

 fected state as a time-keeper. Invention had succeeded in giving to the 

 rising pointer a regular motion, and adapting it well to its purpose. 

 Other advances were made in it, and of these it remains to speak. 

 Improvement, handicapped by the clumsy Roman hours, found in this 

 fact a stimulus to ingenuity. To adapt it to indicate these hours one 

 rude scheme was to reduce the capacity of the vessel from which the 

 water Ho wed by coating it with wax in the winter time. The orifice 

 remaining unchanged it emptied more quickly. The wax was gradually 

 removed as the days lengthened. Of course the same instrument 

 could not serve for both day and night. Less clumsy means for regu- 

 lating the flow, as by adjusting the size of the orifice, were afterwards 

 invented. One of these involved the pass.age of the water through a 

 hollow cone or funnel, in which was an interior cone capable of adjust- 

 ment for each day in the year; another, invented by Ctesibins, left the 

 water-flow, and consequently the rise and fall of the fioat — constant, 

 but included an automatic device by which the graduated scale over 

 wiiich the marker travelled was changed daily. 



This difficulty in adapting tbe clepsydra to keep Roman time is pre- 

 cisely the same that the early Dutch navigators met with on their in- 

 troduction of the clock into Japan, where the division of the day is 

 into ten hours of varying length. The plan they adopted is a clumsy 

 one, but of the same character as that of Ctesibius, since they did not 

 attempt to alter the rate of the clock, but attached movable indications 

 to the dial so that they might be changed with the season. One of 

 these clocks is in the possession of the Bureau of Education, a gift from 

 the Japanese Government after the Centennial Exposition of 1870. 



