618 EEPOET— 1887. 



The two chimneys were attached together hy two curved metal hands suffi- 

 ciently removed from the flame on either side not to affect it. The attachment of 

 these bands to the lower chimney was adjustable, so that the opening through 

 which the central parts of the tiame were seen might be made larger or smaller. 

 By means of small cylindrical blocks, whose thickness was accurately gauged, 

 the width of the opening might be set either to that at which the light 

 emitted was one candle, or, if a greater or smaller light was desired, a candle and a 

 half, or half a candle. The width of the connecting bands was half that of the 

 tube which surrounds the flame. "NVlien these bands were placed in a plane 

 perpendicular to the bar of a photometer, a point midway between their edges and 

 at half the height of the flame might be taken very approximately as that from 

 which the light radiated. 



The liquid with which the lamp is fed is pentane, obtained in a manner 

 already described from American petroleum. 



3. Second Report of the Committee for invitinfi designs for a good Differential 

 Gravity Meter. — See Reports, p. 41. 



4. Report of the Committee for considering the desirahility of combined 

 action for the purpose of Translation of Foreign Memoirs. — See Re- 

 ports, p. 41. 



5. Contributions to Marine Meteorology from the Scottish Marine Station.^ 

 By Hugh Robert Mill, D.Sc, F.E.S.E., F.C.S. 



Observations have been carried on during three and a half years on the tem- 

 perature and salinity of the sea and its inlets at various points of the coast of Scot- 

 land and on various inland lakes. The general result has been to show a marked 

 difference between the seasonal changes of temperature in sea-water and in the 

 water of inland lakes of equal depth. There is a distinct difference also in the con- 

 ditions of the water on the east and on the west coasts of Scotland. 



In the deep rock-basins and submerged valleys of the west coast a very singular 

 vertical distribution of temperature has been detected and its seasonal changes 

 watched. The Clyde sea-area was selected for particular study, as presenting a 

 variety of natural conditions i-eadily accessible at all seasons. Results obtained 

 there show for (1) the b-ish C/ta^itiel a. umiorm temperature from surface to bottom, 

 changing regularly with the season, but higher all the year round than the mean 

 of the enclosed regions; (2) the deep open baxins in free tidal communication with 

 the ocean resemble the Channel at all depths beneath thirty fathoms. The surface 

 water changes more rapidly in temperature than that below, and hence is warmer 

 in summer and colder in winter than the mass ; (3) the deep enclosed basins, almost 

 cut off from the tide and shut in by steep mountain walls, show the greate.st range of 

 annual temperature, and the most complicated vertical distribution. The surface 

 water is quite fresh after heavy rains and freezes in winter. The annual range may 

 be 35° or 40° F., while at the bottom (seventy fathoms) 5° is the greatest range 

 observed, and the maximum temperature there occurs in early spring, when the 

 surface water is at its minimum ; the minimum at the bottom occurs in the begin- 

 ning of autumn, when the surface attains a maximum. 



Superposed layers of water at various temperatures have been frequently 

 observed, and the curves of vertical change show abrupt transitions, often amounting 

 to several degrees in a single foot, at considerable depths beneath the surface. The 

 subject is being investigated from the side of the specific heat and conductivity of 

 sea-water of various salinities. 



' Published in extenso in the Scottish Meteorological Society's Journal, 3rd ser. 

 vol. viii. 



