were taken at the same place and time, and by the same person, Mr. Henry Dean, as those of the previous year. 70 July Aug Sep Oct Nov Dec Jan Feb Mar April May June July Aug 60 40 30 RIVER NIDD. Observations taken by Mr. G. Paul, at Knaresborough. The weekly means from which the above table is calculated are given in Curve No. IX. Mr. Paul's observations are remarkably regalar and CURVE IX.-Nidd, Knaresborough. 9.30 A.M. Water Mar Apr May June July Aug Sept Oct Nov Dec Jan Feb Mar Apr May June July Aug Sep Oct Nov Dec Jan Feb Mar Apr 2 consistently careful. He has reinforced them with some interesting observations on the exceptionally severe winter of 1890-91, which 1 contain some apparently new observations on the protective character of a sheet of ice against the cooling of water by radiation. The rapid fall of temperature following a thaw is analogous to the effect observed on earth thermometers not far from the surface in like conditions. Notes on the River Temperatures for the Winter 1890-91 in the Nidd. By Mr. G. PAUL, Knaresborough. Some very interesting effects were observed during the long period of intense cold. The river was frozen over, but the water under the ice remained at the constant temperature of 34°0 from December 20 to January 2 inclusive. It fell to 33° on thawing on January 3. On January 6 a second period of ice on the water commenced, and during it the temperature under the ice remained at 33°0. Not until the next thaw set in and a third frozen period commenced on February 27, did the temperature of the water under the ice fall to 32°.0. Compared with this the record of earth-temperature at the depth of 1 foot acquires a special interest. An ice-cap was formed on the ground on December 21 and remained until January 21. During this entire month the temperature at the depth of 1 foot scarcely varied. On December 20, before the severe cold set in, the temperature at that depth was 37°4, on the 21st it fell to 36°.5, next day to 36°4, and until January 4 its range was only between 36°4 and 36°3. From January 2 to 17 the temperature at 1 foot was 36°3 or 36°2; on the 18th it fell to 36° 0, and on that day the grass thermometer registered only 3°.8, the minimum temperature of the winter. On January 20 the 1-foot earth thermometer registered 36°4, but on the 21st it fell to 35°5, coincident with a rise of air-temperature and a general thaw. Next day the temperature at 1 foot had risen to 36°.2, and did not again fall below this value. The rest of the observations are given in the form of tabulated monthly means and of curves expressing the weekly means. Had time and other circumstances permitted, many or all of these sets of observations would have been fully discussed, but the mere record suggests many interesting relations as to the period of maximum and minimum temperature, the manner in which water-temperature follows air-temperature, and the effect of situation in latitude and altitude on the rate and amount of monthly change. BRISTOL CHANNEL AND TAFF. The Cardiff Naturalists' Society arranged for three sets of observations from February 1889 to July 1891. One was taken by Mr. Pettigrew in the Cardiff Castle Gardeus on the Dock Feeder, a large stream diverted from the Taff. This set of observations (Curve XII.) shows the water-temperature to be always higher than that of the air. This is the only case in which this relation, so clearly shown in the Cherwell observations, was distinctly seen in another river. The Breaksea Lightship is anchored in the Bristol Channel about the centre, south of Barry Island. Observations were made on it daily at 9 A.M. by Mr. J. Walters and Mr. J. R. Johnson. The record (Curve X.) shows that while the temperature is rising, from February to July, the water is colder than the air; but when the temperature is falling, from |