PHYSICS. 359 



minimum wave length assignable to the minimum ordinate of the heat 

 curve in the spectrum of a source whose temperature varies from 100^ 

 to 0° Centigrade, is a little less than 5,a and a little over G/j." In other 

 words, that " some of the heat radiated by the soil has probably a wave 

 length of over 150,000 of Augstrom's scale, or about twenty times the 

 wave length of the lowest visible line in the solar spectrum as known 

 to Fraunhofer." This research, therefore, has rendered probable the 

 existence of measurable wave lengths of something greater than one 

 two-thousandths of an inch and has proved that the heat radiated from 

 the soil is of an almost totally different quality from that which is re- 

 ceived from the sun. (Am. J. Sci., Jan., 1886, III, xxxi, 1 ; Phil. Mag., 

 May, 1886, V, xxi, 394; Ann. Chim. Phys., December, 1886, VI, ix, 433.) 



In a subsequent paper on hitherto unrecognized wave lengths Lang- 

 ley gives an extended description of the method employed by him for 

 measuring these wave lengths, together with an account of the appa- 

 ratus used in the research. " Broadly speaking," he says, " we have 

 learned through the present measures with certainty of wave lengths 

 greater than 0.005°^'", and have grounds for estimating that we have 

 recognized radiations whose wave length exceeds 0.03^""^, so that while 

 we have directly measured to nearly eight times the wave length 

 known to Newton, we have probable indications of wave lengths far 

 greater, and the gulf between the shortest vibration of sound and the 

 longest known vibration of the tether, is now in some measure bridged 

 over." (Am. J. Sci., August, 1886, III, xxxii, 83.) 



Wiedemann and Leudecking have studied the heat changes which 

 accompany the hydration and solution of colloids. When water is 

 progressively added to colloids such as gelatine, gum arable, gum trag- 

 acaiith, dextrine, starch, etc., two stages of the action are observed. In 

 the first these colloids become hydrated with an evolution of heat. In 

 the second the hydrates formed dissolve in the excess of water with 

 absorption of heat. (Wied. Ann., xxv, 145; J. Phys., Nov., 1886, II, 

 V, 495.) 



Pickering has pointed out some of the sources of error incident to 

 calori metrical work. He finds that the presence of anything but air 

 between the calorimeter and the jacket is most injurious; the space 

 should be entirely open and no cover of any sort should be used. Be- 

 fore reading the thermometers the top of the stem should be tapped 

 for some time, otherwise the mercury lags behind the true temperature. 

 Moreover a thermometer when rising is invariably too low, and when 

 falling is invariably too high. The error thence arising is avoided by 

 conducting the entire experiment either with a rising or with a falling- 

 thermometer. In his experiments the thermometers used had a range 

 of 15° and a total length of 600™"'. The readings were made at tem- 

 peratures between — 1° and 26°, by the following method: The ther- 

 mometer was first heated to the highest temperature required in the 

 experiment, and, by the application of a flame to the mercury column 



