S31 



TRANSPIRATION. 



TRANSPIRATION. 



annihilated ; an effect is produced on the medium, whether it be that 

 the temperature is raised, or that chemical changes are produced, or 

 that the medium is made to emit light of a different kind, as in the 

 rfrnKTwrn of phosphorescence and fluorescence, [ KLUOBMCBXCII.] 

 These phenomena, and especially perhaps the but, indicate that the 

 molecules of the medium are thrown into a state of agHation ; and thus 

 we are led to suppose that the undulations of the luminiferoua ether 

 are spent in producing agitations among the ultimate molecule* of the 

 absorbing body, the consideration of which therefore must form an 

 (tr*i<ol (art of a complete explanation of absorption. That the period 

 of the incident undulation* should play auch an important part in the 

 phenomena, may be illustrated to a certain extent by considering the 

 effect of a series of slight pushes, periodically applied to a body capable 

 of swinging as a pendulum, which will throw it into a state of con- 

 siderable vibration, provided the period of the pushes nearly gree 

 with that of the natural vibration* of the body. If the cause of 

 absorption be that just explained, we must attribute transparency to 

 the existence of such a constitution in the body, that the ether, or a 

 portion of it at least, can in its undulations glide freely among the 

 molecules of the body, without throwing them into a state of agitation. 

 TRANSPIRATION, a term applied by Mr. Graham to a peculiar 

 and fundamental property of the gaseous form of matter in passing 

 through capillary tubes. It differ* from EFFUSION by which gases 

 pass through a small aperture about ii,th of an inch in diauu-ter, into 

 a vacuum ; but some of the results of transpiration correspond with 

 the flow of liquid* through capillary tubes referred to under DIFFU- 

 SION. 



The result* obtained by Dr. Poiseuille, and confirmed by Itegnault, 

 were obtained by sending a liquid under examination through a 

 capillary tube under the influence of condensed air of known preamre. 

 For a minute account of these experiments we must refer to the 

 ' Annalea de Chimie,' 3e aerie, xxi. ; an abstract of them is also given 

 in Professor Miller'* ' Chemical Physics,' where the apparatus is also 

 figured. Among the general results obtained, it appears that the rate 

 of efflux of the liquid when the tube exceeds a certain length (which 

 is greater as the diameter increases) increases directly as the pressure, 

 so that by doubling the pressure, the amount of liquid discharged is 

 double, the times being equal. With tube* of equal diameter the 

 quantities discharged in equal times are inversely a* the length of the 

 tube, so that a tube two inches long discharging 100 grains of liquid in 

 five minutes, a similar tube four inches long would discharge only 50 

 grain* in the same time. In tubes of equal lengths but of different 

 diameters, the flow is as the fourth power of the diameters, so that 

 the flow from a tube ^,th of an inch in diameter would be 16 times 

 a* great as from a tube jfoth of an inch in diameter, or as 1* : 2*. The 

 material of the tube does not appear to influence the result ; but the 

 nature of the liquid does so greatly. In most cases the flow of saline 

 solutions was found to be slower than that of distilled water : the 

 alkalies produced this effect. Certain substances appeared to exert no 

 influence, such as nitrate of silver, corrosive sublimate, iodide of 

 sodium, iodide of iron, nitric, hydriodio, bromic, and hydro bromic 

 acid*. The presence of some other substance* increased the rapidity 

 of the flow, such as hydro-sulphuric acid and hydro-cyanic acid ; the 

 nitrates of potash and ammonia, the chloride* of potassium and 

 ammonium ; the iodide, bromide, and cyanide of potassium. A slight 

 increase in temperature generally augments the flow; water at 113 

 escaping 2} times quicker through the same tube than it did at 41. 

 But it is remarkable that concentrated solutions of iodide of 'potassium 

 above 140 r'ahr., and of nitrate of potash above 104 flow more slowly 

 than distilled water. In general, however, the solutions contained 

 only 1 per cent, of the substances ; and they were exposed to a 

 pressure equal to that of a column of water 1 metre (39-37 inches) 

 high, at the temperature of 52-16, and escaped through a tube 64 

 millimetres (2-519 inches) in length, and 0-24946 millimetres (0-0108 

 in.'!., in diameter. No connection has been traced between the rate of 

 efflux of the liquid and its density, capillarity or fluidity. The dilu- 

 tion of alcohol retards its efflux, up to a certain point, beyond which it 

 increase* it, the minimum efflux corresponding with that mixture of 

 alcohol and water which i* attended with the maximum contraction. 

 The solubility of a substance in water exerts only a secondary influence 

 on the efflux. Dr. Poiseuille ha* shown that variou* solutions intro- 

 duced into the blood of a living animal apparently produce effects of 

 acceleration or retardation on the capillary circulation corresponding 

 with those noticed with the same liquids in gbus capillary tubes. 



Substituting gases for liquids, it appears that the rate of efflux, or 

 the velocity of transpiration for each gas, is independent of its rate of 

 diffusion. In Graham's experiments on this subject, (' Phil. Trans.' 

 1840 and 1849) the gas was contained in a graduated jar, standing over 

 water, and mtpfU'W' so that the water on the inside should be kept on 

 the same level as that on the outside. On allowing the jar to sink, the 

 gas wa* expelled by a flexible tube into a bent tube containing chloride 

 of calcium, and being thus dried, it passed through a long capillary 

 tube, and thus entered the exhausted receiver of an air pump, which 

 was either kept exhausted, or the amount of exhaustion was noted 

 by means of the gauge, the quantity of gas that' entered the receiver 

 in a given time being carefully noted. 



By employing a certain length of tube increasing with the diameter, 

 not the same for all gases, it appears that the rate of transpiration 



increase* directly as the pressure, so that equal volumes of air at 

 different densities require times inversely proportioned to the densities : 

 thus a pint of air double the density of the atmosphere will pass 

 through the capillary tube into vacuum in half the time required for 

 a pint of air of ordinary density. With tubes of equal bore the 

 volume transpired in equal times is inversely as the length of the tube ; 

 thus, if 30 cubic inches of gas were transpired through a tube 10 feet 

 long in 5 minutes, a similar tube 20 feet long would only allow the 

 passage of 15 cubic inches in the same time. It was also found that 

 the transpiration of-equal volumes becomes slower as the temperature 

 rises. Uniform results were also obtained, whether the tubes were of 

 copper or of glass, or a porous mass of stucco were employed. Trans- 

 piration was found to vary with the chemical nature of the gases. The 

 velocities of transpiration of different gases bora a constant relation to 

 each other, independently of their densities, and it was thought 

 probable that the rate of transpiration is the resultant of a kind of 

 elasticity depending on the absolute quantity f heat, latent as well aa 

 sensible, which different gases contain under the same volume, so that 

 transpiration seems to be Ultimately connected with the specific heat 



1 I :-' - 



Oxygen has apparently the slowest rate of transpiration, and is taken 



JB the unit in the following table. It is found that a mixture of 

 equal volumes of two gases does not always give the mean tranxpira- 



bility. Thus the time for the transpiration of hydrogen is much 

 prolonged by admixture of oxygen, the rate being 0-9008 instead of the 

 mean 0'72. 



Times for 



Transpiration of Velocity of 



Gates. equal volumes. Transpiration. 



Oxygen . ... 1-0000 1-0000 



Air 0-9030 1-1074 



( Nitrogen 08768 1-141 



{ Binoxide of oitrogsn . . .0-8764 1-141 



( Carbonic oxide , . . .0-8737 1-114 



[ I'rotuxide of nitrogen . . .0-7493 1-334 



{ Hydrochloric acid . . .0-7363 1-361 



I Carbonic acid 0-7100 1-369 



Chlorine 0-6664 1'500 



Sulphurous acid . . . . 0-6500 1 5.18 



Sulphuretted hydrogen . .0-6195 1-614 



Light carburctted hydrogen . .0-5510 1-815 



Ammonia 0-5115 1-9S* 



Cyanogen 0-5080 1-978 



OleBantgas . . . .0-5051 1-980 



Hydrogen 0-4370 l*Mf 



Vapours. 



Bromine, about . 

 Sulphuric acid, anhydrous 

 Bisulphide of carbon . 

 Chloride of methyl . 

 Chloride of ethyl . . 

 Oxide of methyl 

 Hydrocyanic acid . 

 Ether .... 



1-0000 

 1-0000 

 0-6195 

 0-5475 

 0-4988 

 0-4826 

 0-4600 

 04400 



In these experiments capillary glass tubes, varying from 20 feet to 

 2 inches in length, gave similar results, where a sufficient resistance 

 was offered to the passage of the gas. The effusion of gases, or the 

 discharge by an aperture in a thin plate, is dependent in all gases upon 

 a constant function of their specific gravity ; but the discharge of the 

 same gases from tubes has no uniform relation to the density of the 

 gases. Both hydrogen and carbonic acid, for example, pass more 

 quickly through a tube than oxygen, although the one is lighter and 

 the other heavier than that gas. 



One of the capillary tubes used by Graham was as much as 22 feet 

 in length ; it was made up of several portions of capillary tube as 

 nearly equal in bore as could be judged of by the eye, cemented together 

 by the blow-pipe so as to form a continuous length, but bent up into 

 coils for the convenience of using. It* extremities were connected 

 with block-tin tubes, proceeding from the drying-tube and air-pump 

 jar by means of thick caoutchouc adopters. This long capillary tulio 

 allowed one cubic inch of air to pass into a vacuum in 15-04 st>> 

 two inches of the tube held 2'65 grains of mercury, which gives a 

 diameter of 0-0222 inch, or ,',Ui of an inch. 



It was found possible to form a capillary tube of copper of less dia- 

 meter than one of glass, by the following means : o cylindrical boh- 

 was first drilled in the axi* of a solid copper rod, four or five inches 

 long, which rod was then extended by passing it through a wire draw- 

 plate. An iron wire, or triplet, wa* placed within the copper tube, and 

 drawn through the plate at the same time, in order to keep the surface 

 of the copper tube smooth and uniform. It was found necessary to 

 pull out the iron wire every time the copper was drawn, to prevent its 

 becoming fixed. The iron wire was then extended separately, and 

 again introduced into the copper tube, and the operation of drawing 

 out was repeated. In this way the copper tube was extended 11 feet 

 8 inches, and it remained perfectly sound and air-tight. One cubic 

 inch of air passed through it into a vacuum in 22-12 seconds. Its 

 diameter was thus found : Of the iron wire on which the copper wa* 

 hut drawn, 927 inches weighed 18-30 grains, or one inch weighed 

 0-19/4 grains. Taking the specific gravity of iron at 77, this, gives the 



