60 FLUIDITY AND PLASTICITY TABLE XXL—VISCOSITY OP AIR AT 15° Method 9 Observer Transpiration ............... 5,507 Breitenbach (1899) Transpiration ............... . 5,502 Schultze (1901) Transpiration 5 528 Markowski (1904) Transpiration 5,502 Schrnitt (1909) Transpiration 5.531 Knudsen (1909) ing data might be improved by a critical study for the purpose of making the needed corrections. It is a curious fact that the kinetic energy correction has been so little understood and appre- ciated. Even in the case of liquids it is very commonly neglected although it may amount to several per cent of the viscosity to be measured. It is sometimes stated that no kinetic energy correction is necessary when liquids now through a capillary from one reservoir to another and not into the air. The Ostwald viscorneter (cf. p. 75), is used more than any other but it appears that no kinetic energy correction is ever applied. It is true that the instrument is used for relative measurements only, but this fact does not cause this correction to be without effect in the cal- culation, for the reason that the kinetic energy correction is not proportional to the viscosity. There may be those who would maintain that the correction for kinetic energy, as given above, is not the correct one to apply to gases, cf. Fisher (1909). But that a correction is unnecessary cannot be maintained even in the case of gases, in view of Hoff- mann's work on the interrupted now of gases. One of Ms capil- laries was cut into 28 pieces without loss. In flowing through the interrupted capillary, the kinetic energy correction would be increased twenty-eight fold, as has been already indicated on page 28. As a matter of fact the time of flow was considerably greater in the interrupted capillary, proving the importance of the correction. It would be particularly interesting to see whether an experimental verification of the correction would be obtained by an intensive study of the data of Hoffmann (1884). The term "specific viscosity" has been very largely used and may here receive brief comment. Water at 0° has been taken as a standard with a specific viscosity of 100, but water at 25°