The London gardening weeklies, during the late severe winter, seldom issued a copy without containing some warning, complaint, or instruction in regard to Thermometers. The low readings recorded in some of the ordinarily favoured counties were received with reservation, and the untrustworthiness of instruments was accredited with the extreme temperature indicated. Thermometers may be corrected as follows:-

The common and misleading idea that "freezing-point," as marked on the scale of the Thermometer, is the point at which water freezes is an error. Under different circumstances, which it is not our present purpose to explain, water freezes at different temperatures below what Thermometer - makers understand as "freezing-point," of course, or exactly at it. But while the air may sink below freezing-point before water freezes, the person who watches the first indication of ice to know whether his Thermometer be correct or not, may be led into error; and the air in contact with grass may be freezing, while 3 feet from the ground it may be a good many degrees higher. Evidently, then, the freezing-point cannot be found this way, although many persons are astonished at the supposed vagaries of Thermometers. But ice melts invariably exactly at one point, and no variation occurs in the phenomenon, and the "freezing-point" of the Thermometer-makers is just the point at which ice melts. We corrected four Thermometers lately in this way.

We took a pail with a hole in the bottom, which allowed melting water to escape, and drained it like a pot (a large pot would do) with lumps of ice, which was exposed to the air of a warm house a few minutes to start it a-thawing. Over these we put more ice, pounded small, and buried the Thermometers in the centre - leaving the pail and its contents where the temperature was over 60° for a couple of hours, and then uniced the Thermometers. Before this they were placed side by side, and all indicating somewhat different temperatures. There was only one "cheap" one amongst the lot - the rest were marked "warranted." On picking them out, No. 1 proved two degrees too low; No. 2 proved three degrees too high; No. 3 was ten degrees too high; and the "cheap" unwarranted was correct - at freezing-point. Whether they be correct at any other point could only be tested alongside of one with a scale reaching to boiling-point (at the level of the sea - elevation makes a difference), and by this means proved correct; but, unfortunately, we did not at that time possess one with so extended a scale.

One point more demands attention. Sometimes a portion of the spirit - in spirit instruments - gets detached, and some difficulty is experienced in getting the portion to reunite. Some people recommend waving the instrument smartly at arm's length to secure this. An easier, a better, and a much more scientific way, is simply to plunge it into water of the same temperature as there are degrees in the scale of the instrument. This causes the spirit to fill the tube, when a sudden plunge into cold water will prevent its swelling more, and then it will be found that the column of spirit will recede in a body. A. H.