The thermometers constructed by Messrs. Negretti and Zambra for this purpose do not differ materially from those usually made under the denomination of Sixe's thermometers, except in the following most important particular:--The usual Sixe's thermometers have a central reservoir or cylinder containing alcohol; this reservoir, which is the only portion of the instrument likely to be affected by pressure, has been, in Negretti and Zambra's new instrument, superseded by a strong outer cylinder of glass, containing mercury and rarefied air; by this means the portion of the instrument susceptible of compression, has been so strengthened that no amount of pressure can possibly make the instrument vary. This instrument has been tested in every possible manner, and the results have been highly satisfactory, so much so as to place their reliability beyond any possible doubt. The scales are made of porcelain, and are firmly secured to a back of oak, which holds in a recess the bulb with its protecting shield, and is rounded off so as to fit easily and firmly in a stout cylindrical copper case, in which the thermometer is sent down when sounding (see fig. 69). The lid of the case is made to fit down closely, and water-tight. At the bottom of the case is a valve opening upward; and the lid has a similar valve. These allow the water to pass through the case as the instrument sinks, so that the least amount of obstruction is offered to the descent. At the lower end of the case is a stout brass spring, to protect the instrument from a sudden jar if it should touch the bottom while descending rapidly. As the instrument is drawn up, the valves close with the weight of water upon them, and it arrives at the surface filled with water brought up from its lowest position. The deep-sea thermometers used in the Royal Navy are of this pattern. [Illustration: Fig. 69.] =90. Johnson's Metallic Deep-Sea Thermometer.=--The objection to the employment of mercurial thermometers for ascertaining the temperature of the ocean at depths, arising from the compression of the bulbs, which was of such serious consequence previous to the modification made in the construction of the instrument by Messrs. Negretti and Zambra, led to the construction of a metallic thermometer altogether free from liability of disturbance from compression by the surrounding water; which, however, is certainly not so sensitive to changes of temperature as mercury. This instrument is the invention of Henry Johnson, Esq., F.R.A.S., and is thus described by him:--