On the Motion of Gases .
By THOMAS GRAHAM , Esq. , F.R.S. , Professor of Chemistry in University College , London ; Hon. Fellow of the Royal Society of Edinburgh ; Corresponding Member of the Royal Academies of Sciences of Berlin and Munich , of the National Institute of Washington , &lt; sc .
Received June 18 , -Read June 18 , 1846 .
THE spontaneous intermixture of different gases , and their passage under pressure through apertures in thin plates and by tubes , form a class of phenomena of which the laws have been only partially established by experiment .
The separation of two gases by a porous screen , such as a plate of dry stucco , will prevent for a short time any sensible intermixture arising from slight inequalities of pressure , but such a barrier is readily overcome by the diffusive power of the gases , which is fully equal to their whole elastic force .
Hence a cylindrical glass jar with a stucco top , filled with any gas and standing over water , affords the means of demonstrating the unequal diffusive velocities of air and the gas , by the final contraction or expansion of the gaseous contents of the jar , after the escape of the gas is completed .
Compared with the volume of air which has entered , the volume of gas which has passed simultaneously outwards is found to be in the inverse proportion of the square root of the specific gravity of the gas .
The diffusive velocities therefore of different gases are inversely as the square root of their densities ; or the times of diffusion of equal volumes directly as the square root of the densities of the gases* .
Such is also the theoretical law of the passage of gases into a vacuum , according to the well-known theorem that the molecules of a gas rush into a vacuum with the velocity they would acquire by falling from the summnit of an atmosphere of the gas of the same density throughout ; while the height of such an atmosphere , composed of different gases , is inversely as their specific gravities .
This is a particular case of the general law of the movement of fluids , well-established by observation for liquids , and extended by analogy to gases .
The experiments which have already been made upon air and other gases , by M. P. S. GIRARDt and by Mr. FARADAY+ , are sufficient to show that the discharge of light is more rapid than that of heavy gases ; and are interesting as first approximations , although incomplete and lending a very imperfect support to the theoretical law .
Indeed some results obtained by these experimenters and others , appear wholly inconsistent with that law , such as Mr. FARADAY'S CUrioUS observations of the change of the relative rates of hydrogen and olefiant gases in passing through a capillary tube under different pressures ; and my own observation , that carbonic acid gas is forced by pressure through a porous mass of stucco as quickly or more so than air is , although more than a half heavier ; and that other gases pass in times which have no obvious relation to their diffusive velocities* .
In studying this subject , I found that it was necessary to keep entitely apart the two cases of the passage of a gas through a small aperture in a thin plate and its passage through a tube of sensible length .
The phenomena of the first class then became well-defined and simple , and quite agreeable to theory .
Those of the second class also attained a high degree of rtegularity , where the tubes were of great length , or being short were of extremely small diameter .
Capillary glass tubes , which varied in length from twenty feet to two inches , were found equally available and gave similar results , where a sufficient resistance was offered to the passage of the gas .
The rate of discharge of different gases from capillary tubes appears to be independent of the nature of the material of the tube , in so far as the rates were found to be similar for tubes of glass and copper , and even for a porous mass of stucco .
But while the discharge by apertures in thin plates is found to be dependent in all gases upon a constant function of their specific gravity , the discharge of the same gases from tubes has no uniform relation to the density of the gases .
Both hydrogen and carbonic acid , for instance , pass more quickly through a tube than oxygen , although the one is lighter and the other heavier than that gas .
I shall assume then for the present , that in the passage of gases thr ?
ough tubes we have the interference of a new and peculiar property of gases ; and on the ground of a radical difference in agency speak of the two classes of phenomena under different names .
The passage of gases into a vacuum through an aperture in a thin plate I shall refer to as the Effusion of gases , and to their passage through a tube as the Transpiration of gases .
The determination of the coefficients of effusion and transpiration of various gases will be the principal object of the following paper .
PART I.-EFFUSION OF GASES .
1 .
Effusion into a Vacuum by a glassjet .
The glass jet was formed from a short piece of a capillary thermometer tube , of which the bore was cylindrical , to which a conical termination was given by drawing it out when softened by heat and breaking the point .
The aperture at the point of the jet was cylindrical , in a flat surface , and so small that it could only be seen distinctly by means of a magnifying-glass ; its size , compared with other apertures , may be expressed by the statement that one cubic inch of air of the usual tension passed into a vacuum through this aperture in 2'18 seconds .
By means of a perforated cork this glass jet was fixed within a block-tin tube , through which the gas was to be drawn ; with the point of the tube dirlected towards the magazine of gas , so that the gas in passing towards the vacuum entered the conical point of the jet instead of issuing from it .
This form of the aperture reduced the rubbing surface of glass to a thin ring , or made it equivalent to an aperture in a very thin plate ; but the mode of placing the jet , or direction in which the current passed through the aperture , was found afterwards to be of little consequence .
The gas for an experiment was contained in a glass jar , of an elliptical form , balanced like a gasometer over water , and terminated at top and bottom with two short hollow cylindrical axes , of an inch in diameter ; its capacity between two marks , one on each of the cylindrical ends , being 227 cubic inches .
From this gasometer the gas was conveyed directly into a U-shaped drying tube , 18 inches in length and 0*8 inch in diameter , filled in some cases with fiagments of chloride of calcium , in others with fiagments of pumice-stone soaked in oil of vitriol ; the putnice , when used , having been first washed with water , to deprive it of soluble chlorides .
From the drying tube , the gas entered the tin tube occupied by the glass jet , one end of that tube being connected with the drying tube , and the other with an exhausted receiver on the plate of an air-pump .
The apparatus described is exhibited in fig. 1 of Plate XXXIII .
, with the exception of the elliptical gasometer , the place of which is occupied there by the counterpoised jar A in the water-trough .
The gas was thus forced through the minute aperture by the whole atmospheric pressure .
In making an experiment with any other gas than atmospheric air , a considerable quantity of the gas was first blown through the drying tube , from the gasometer , to displace the air in the former ; and to do this quickly an opening was made into the air-channel beyond the drying tube , at G , by which gas might be allowed to escape into the atmosphere without proceeding further or being drawn through the glass aperture into the vacuum .
This side aperture was closed by a brass screw and leather washer .
In making an experiment , the gasometer was filled with the gas to be effused , and then connected with the air-pump receiver , in which a constant degree of exhaustion was maintained by continued pumping .
The interval of time was noted in seconds , which was required for the passage of a constant volume of gas , amounting to 227 cubic inches , namely , that contained between the two marks in the elliptical gasometer .
Or , the volume of gas effused was more strictly 227 cubic inches , minus the volume of aqueous vapour which saturates air at the temperature of the experiment ; the vapour being withdrawn from the gas , after it left the elliptical measure and before it reached the effusion aperture .
It is scarcely necessary to add that great care is necessary during these and all other experiments on gases , to maintain a uniform temperature .
The use of a fire or stove in the room in which the experiments were conducted was therefore avoided , and such arrangements made that the temperature was kept for five or six hours within a range of a single degree of FAHRENHEIT'S scale .
Hydrogen .
In the experiments first made with air and hydrogen the temperature 575 was 590 FAHR. , and the height of the barometer 30 14 inches ; a uniform exhaustion was maintained in the air-pump receiver of 29 3 inches , as observed by the gauge barometer attached .
tI'he constant volume of dry air passed into the vacuum , or was effused , in three experiments , in 494 , 495 , and again in 495 seconds .
The constant volume of dry hydrogen was effused , in two experiments , in 137 and again in 137 seconds .
Calculating from 495 seconds as the time for air , we haveTime of effusion of air ... ... .
1 Time of effusion of hydrogen ... .
0277 Or , the result may be otherwise expressed , taking the reciprocals of the last numbers : Velocity of effusion of air ... 1 Velocity of effusion of hydrogen ... 3*613 The specific gravity of hydrogen gas , according to the most recent and exact determnination , that of REGNAULT is 0'06926 , referred to air as unity ; of which the square root is 0'2632 , and the reciprocal of the square root 3'7994 ; to which the numbers for the time and velocity of hydrogen above certainly approximate .
Oxygen and Nitrogen.-Temperature 60 ?
; exhaustion maintained at 29,3 inches .
The constant volume of air was effused in 494 seconds , of oxygen in 520 seconds , and of nitrogen in 486 seconds , in one experiment made upon each gas .
Hence the following results : Time of effusion .
Square root of Veliocity of usiosquare density .
root of density .
Air ... ... ... ... ... ... ... 111 Oxygen ... ... ... ... ... ... ... ... 1-053 1'0515 0o9500 0'9510 Nitrogen ... ... ... ... ... ... ... . .
0-984 0*9856 1-0164 1-0146 The densities made use of are those of M. REGNAULT , namely , 1#10563 for oxygen , and 0'97137 for nitrogen .
It will be observed , that the times of effusion of these two gases correspond as closely with the square roots of their densities as the mode of observation will admit of ; the times observed being within one second of the theoretical times .
Carbonic Oxide.--This gas was prepared by the action of oil of vitriol upon pure crystallized oxalic acid , and subsequent washing with alkali .
The temperature during the effusive experiment was 60 ?
3 ; the usual exhaustion was maintained .
The time of effusion of air was 494 seconds ; of the same volume of carbonic oxide 488 seconds : 576 The effusion-rate of this gas approaches therefore very closely to the theoretical number .
In the calculations the density of carbonic oxide is taken at 0'96779 , as found by WREDE .
Carburetted Hydrogen .
, CH.--This was the gas of the acetates , prepared by heating a mixture of acetate of soda with dry hydrate of potash and lime .
The temperature of the gases effused being 59 ?
'5 , and the exhaustion 29*3 inches ; the constant volume of air passed through the aperture in 493 seconds , of carburetted hydrogen in 373 seconds : Time , air = 1 .
Theoretical time .
Velocity .
Theoretical velocity .
Carburetted hydrogen ... ... 0-756 0-7449 1'322 1-3424 The density of carburetted hydrogen is taken at 0'5549 in the calculations .
Carbonic Acid and Nitrous Oxide .
In the first experiment with carbonic acid , the gasometer with the gas was floated as usual over water ; thermometer 580 ?
5 .
The effusion of air took place in 495 seconds , of carbonic acid in 595 seconds .
To diminish the loss of the latter gas occasioned by its solubility in water , a second experiment was made over brine : the time required by the carbonic acid was now 603 seconds .
The velocity of effusion of carbonic acid is by the first experiment 0'832 ; by the second it approaches more nearly the theoretical number , calculated from 1*52901 ( REGNAULT ) as the density of this gas , as appears below : Time , air =1 .
Theoretical time .
Velocity .
elit Carbonic acid ... ... ... ... ... ... 1*218 1-2365 0*821 08087 The observation on nitrous oxide was made on a different occasion , with a temperature of 62 ?
'5 .
The time of effusion of air was then 488 seconds ; of nitrous oxide 585 seconds , the gas being collected over water : Time , air =1 .
Theoretical time .
Velocity .
Theoretical velocity .
Nitrous oxide ... ... ... ... ... .
1199 1'2365 0-834 0'8087 The specific gravity of nitrous oxide is assumed in the calculations to be the same as that of carbonic acid .
The time of effusion of both of these gases is shortened by the loss of a portion of the gas , by solution in the water of the pneumatic trough during the period of the experiment , and falls below the theoretical number .
In carbonic acid over brine , where the injury is least from this cause , the observed velocity is , however , still within one-seventieth part of that calculated from the specific gravity of the gas .
Olefiant Gas.-When this gas is prepared by heating sulphuric acid , of specific gravity 1'6 , with strong alcohol at the temperature of 320 ?
, in the proportion of six parts of the former to one of the latter , it appears to come off at first very pure , as it is entirely absorbed by the perchloride of antimony , and contains therefole no carbonic oxide .
But it is really contaminated , I find , by a portion of another heavier gas or vapour ( not ether vapour ) , which cannot be entirely removed from it by washing with alkaline water , oil of vitiriol , or strong alcohol , and which may raise the density of the gas above that of air .
As the evolution of gas proceeds , the proportion of the heavy compound diminishes , and it finally disappears , and the gas attains its theoretical density ; but it is then again contaminated with more or less carbonic oxide .
The latter gas , however , being of sensibly the same density as olefiant gas , is not likely to exert any influence upon its effusion rate .
But before these facts were ascertained this jet became unserviceable from an accident , and the experiments made with it were all made upon the dense olefiant gas , and gave an effusion time which slightly exceeded that of air .
2 .
Effusion into a Vacuum by a perforated brass plate A. A minute circular aperture was made by means of a fine drill in a thin plate of sheet brass -1-th of an inch thick , and the opening still further diminished by blows from a small hammer , of which the surface was rounded .
A small disc of the brass plate was then punched out , having the aperture in the centre , which was soldered upon the end of a short piece of brass tube , of quill size , so as to close the end of the cylinder .
This brass tube was then fixed , by means of a perforated cork , within the tin tube , used as formerly , for conveying the gas from the gasometer jar to the air-pump receiver ; so that the gas should necessarily flow through the small aperture in its passage , as before through the glass jet .
The aperture was of an irregular triangular form , in consequence of the hammering of the plate .
One cubic inch of air of usual tension passed into a vacuum through this aperture in 12'56 seconds .
The volume of gas effused in an experiment was the same as before , and the other arrangements similar , but the aperture in the brass plate being smaller than that of the glass jet , the effusion was considerably slower .
The constant volume of 227 cubic inches of the following gases passed into a vacuum of 29*3 inches by the attached inercurial gauge , at the tenlperature of 63 ?
-3 , in the following times:(1 .
) Air in 47 ' 32 " , or 2852 seconds .
( 2 .
) Nitrogen in 46 ' 47 " , or 2807 seconds .
( 3 .
) Oxygen in 50 ' 1 " , or 3001 seconds .
( 4 .
) Hydrogen in 13 ' 8 " , or 788 seconds .
( 5 .
) Carbonic acid ( over brine ) in 56 ' 54 " , or 3414 seconds .
578 These results , referred to air as unity , are as follows : Time of effusion .
Theoretical time .
Velocity of Theoretical effusion .
velocity .
Air ... ... ... ... ... ... ... ... ... ... 111 Nitrogen ... ... ... ... ... ... ... . .
0-9842 0-9856 1i0160 1-0146 Oxygen ... ... ... ... ... ... ... ... 10502 1'0515 0-9503 0'9510 Hydrogen ... ... ... ... ... ... ... 0-2763 0-2632 3-607 3-7994 Carbonic acid ... ... ... ... ... ... 11971 1-2365 0-8354 0-8087 T'he experimental results of the velocity of eff'usion of nitrogen and oxygen accord very closely with theory , the velocity of the first being only 0-0014 in excess , and the second 0'0007 in deficiency .
Indeed the differences fall within the unavoidable errors of observation in determining the specific gravity of these gases , unless conducted with the greatest precautions .
Of hydrogen , the velocity of effusion observed is 3-607 times instead of 380O times greater than air .
It thus suffers a small but sensible reduction of its velocity , which can be referred , as will afterwards appear , to the thickness of the plate and the aperture being in consequence sensibly tubular .
A portion of the carbonic acid gas must have been absorbed by the brine dutring the long continuance of the experimrent , nearly an hour ; to which the quickness of the rate of that gas may be referred ; the velocity of its passage being thus apparently increased from 08 l to 0'835 .
The experiment was varied by observing the time in which gas entered a vacuous receiver upon the plate of the air-pump , in quantity sufficient to depress the gauge barometer from 28 to 23 inches .
An exhaustion was always made at first of upwards of 29 inches , and the instant noted at which the mercury passed the 28th and 23rd inches of the scale .
The times of effusion were as follows , the temperature being 66 ? .
Experiments .
Velocity of effusion .
1 .
2 .
3 .
Mean .
i Observed .
Calculated .
Air ... ... ... ... ... ... ... ... ... .
474 474 ... ... ... . .
474 11 Oxygen ... ... ... ... ... ... ... ... 501 502 499 I 510 Oxygen .
501 502 499 500'7 ' 0-9467 0'9510 Nitrogen ... ... ... ... ... .
468 469 ... ... ... . .
468-5 1-0117 1-0146 Olefiant gas ... ... ... ... . .
467 469 ... ... ... ... 468 1-0147 Carburetted iydrogen ... ... 357337 1 13278 13369 Carbonic acid ... ... ... ... 573 573 ... ... ... . .
573 0-8272 08087 The same close correspondence is manifest here between the observed and calculated velocities .
The whole results leave no doubt of the truth of the general law , that different gases pass through minute apertures into a vacuum in times which are as the square roots of their respective specjfic gravities ; or with velocities which are inversely as the square roots of their specific gravities ; that is , according to the same law as gases diffuse into each other .
It appears that the proper effect of effusion can only be brought out in a perfect 4 F2 579 manner when the gas passes through an aperture in a plate of no sensible thickness , for when the opening becomes a tube , however short , the effluent gas meets a new resistance which varies in the different gases according to an entirely different law from their rates of effusion , namely , the resistance of transpiration .
The deviation is most considerable in hydrogen , which rapidly loses velocity if carried through a tubular opening , when compared with air .
This was illustrated by experiments made upon the glass jet of the former observations ; which was operated upon in four different conditions as to length .
The point had been drawn out rather long at first , so that it admitted of portions of 0*2 inch , 0'1 inch , and 0'07 inch , being broken off successively before it was reduced to the form of a blunt cone , which it had when used in the experiments already detailed .
Air and hydrogen were effused from this jet into the exhausted receiver till the mercurial gauge fell from 28 to 4 inches , with the jet in the different states described .
When the glass jet was of greatest length , the time of air was 335 and 337 seconds in two experiments , and of hydrogen 120 seconds in two experiments ; which give 2'800 as the velocity of effusion of hydrogen .
After the first portion was broken from the point , by which of course the aperture was enlarged , the time of air was 175 seconds in two experiments , of hydrogen 55 and 56 seconds ; giving 3*153 for the effusive velocity of hydrogen .
After the second abridgement in its length , the time of passage of air by the jet was 110 seconds in two experiments , of hydrogen 33 , 32 and 33 seconds in three experiments ; giving 3*33 for the velocity of hydrogen .
When still further reduced in length , a larger jar being used as the vacuous receiver , the time of air was in two experiments 408 and 410 seconds ; of hydrogen in three experiments , 122 , 120 and 122 seconds , giving 3X38 for the velocity of hydrogen .
Thus , as the jet was progressively shortened , the relative velocity of the passage of hydrogen continually rose , passing through the numbers 2'8 , 3'153 , 3*33 and 3-38 .
By reversing the direction of the stream of gas through the aperture in its last condition , the effect of friction was still further diminished , and the velocity of hydrogen raised to 3'61 , as in the experiments previously recorded , which were iade with this jet in an inverted position .
It may be fairly presumed , therefore , that if the length of the tube or thickness of the plate containing the aperture was still further diminished , the effusive velocity of hydrogen , compared with air , would be increased , and approximate more nearly to 3*80 , the theoretical number .
The tubularity of the opening quickens , on the contrary , the passage of carbonic acid and nitrous oxide in reference to air ; for these gases are more transpirable than air , although less effusive : hence their observed time of effusion is always sensibly less than their calculated time .
580 Effusion of Nitrogen and Oxygen , and of mixtures of these Gases under differezt pressures , by a second perforated brass plate B. This brass plate was of the same thickness as the last ( -2-ith of an inch ) ; the aperture was circular and also 218th of an inch in diameter , as measured by a micrometer ; and the velocity with which air of the usual tension passed into a vacuum by the aperture , one cubic inch in 608 seconds .
T'he rate of passage was therefore rather more than twice as quick as by the first perforated plate A. A two-pint jar was used as the air-pump receiver , or aspirator-jar , as it may be called ; and the capacity of the vacuous space into which the gas effuses , including the tubes and channels of the air-pump as well as the jar , was found to be 72'54 cub. in .
An exhaustion was always first made of about 29 inches by the gauge barometer of the pump , and then the gas allowed to enter from a counterpoised bell-jar over water ( fig. 1 , Plate XXXIII .
) .
The instant was noted at which the mercury fell to 28 inches , when the observation began , and again at 20 and 12 inches , or after two intervals of 8 inches each ; and again at 4 and 2 inches by the gauge barometer .
The experiments were made successivelyon the same dayin the order given , with the barometer at 29*34 inches and thermometer at 49 ? .
A small thermometer placed within the aspirator-jar was observed to rise 1 ?
FAHR. very uniformly during the continuance of an experiment .
The effusion of air is repeated at the close of the experiments to determine whether or not any change of rate had occurred during their continuance .
TABLE I.-Effusion .
Mixture of Gauge barometer in Air .
Ntrogen .
Oxygen 50 nitrogen + 50 oxygen .
inches. .
I I. .
| .
II .
, , , ' , , , ... / * 28 00000000 20 120 120 119 119 126 126 122 122 12 123 122 120 120 128 130 125 125 8 68 68 67 67 72 72 70 69 4 84 84 83 83 89 88 85 86 2 57 57 55 54 59 59 57 57 452 , 451 444 444 474 475 459 459 TABLE 1I.-Effusion .
25 nitrogen + 75 oxygen .
75 nitrogen + 25 oxygen .
Air .
Gauge barometer in inches .
I. | II .
I. II .
28 00000 20 122 122 122 121 120 12 125 125 122 122 123 8 70 69 68 68 68 4 85 86 85 85 83 2 57 57 56 56 57 459 459 453 453 451.~~~~~ ii ii .
581 The near approach to equality in the times from 28 to 20 , and from 20 to 12 inches throughout the whole of these experiments , is very remarkable .
Under an average pressure of 24 inches in the former portion of the scale and of 16 in the latter , the gases effuse with nearly equal velocities ; which confirms the observation of MM .
DE SATNT-VENANT and WANTZEL , on the passage of air through a minute aperture , namely , that above two-fifths of an atmosphere , the further increase of the pressure is attended with a very slight increase in the velocity of passage* .
In the experiments above with an increase of pressure from 16 to 24 , or of one-half , the increase in velocity is not in general more than one-sixtieth part .
In the table which follows the average times are given , which the gauge barometer required to fall from 28 to 12 inches , and from 12 to 4 inches , taken from the preceding table , for air , nitrogen and oxygen , and also the ratio between the times of these gases , that of air being taken as unity , with their relative velocities , also referred to the velocity of air .
Time in seconds .
Time of air = 1 .
Velocity of air = 1 .
Gauge barometer ... ... ... .
: ... Air .
Nitrogen .
xygen .
e Nitrogen .
Oxygen .
Nitrogen .
Oxygen .
From 8 to 12 inches .
; 2425 239 255'0 0'9855 1-0515 1*0146 0 9510 From 12 to 4 inches .
152-0 150 160'5 0'9868 : 1*0558 1-0133 0 9470 These results do not indicate any material difference between the ratios of effusion of these gases at different pressures .
At the low as well as the high pressure , the velocities are in close accordance with the law of effusion ; indeed they correspond as closely as the shortness of the time of observation justifies any inference ; the small deviations observable being quite within the amount of errors of observation .
The i'esults for the mixtures of oxygen and nitrogen are as follows , for similar divisions of the scale : Time in seconds .
Time , air 1 .
Gauge barometer ... ... i* i ... ... . .
I 1 .
Mixture .
11 .
Mixture .
III .
Mixture .
50 N+ 50 0 .
25 + 75 0 .
75 N+ 25 0 .
From 28 to 12 inches .
247 252 243*5 1'0185 1-0391 10041 From 12 to 4 inches .
155 157 153 10197 103 ?
9 1-0066 In these instances , as well as in the unmixed gases , the results do not justify the inference of any difference in the ratios of effusion at low from the ratios which hold at high pressures .
It appears , on comparing the times observed of the mixtures with the times calculated from the unmixed gases , that they sensibly agree .
Thus the mean rate or time of 50 nitrogen + 50 oxygen , or square root of the specific gravity of that mixture , is 1*0191 , the observed rate 1P0185 ; of 25 nitrogen + 75 oxygen , the mean rate is 1P0354 , the observed rate 1*0391 ; of 75 nitrogen + 25 oxygen , the mean rate is 1 0025 , the observed time 1 0041 , in the range between 28 and 12 inches of the gauge barometer .
Lastly , the particular mixture forming atmospheric air has already been seen to have the rate corresponding with its specific gravity or its composition .
It may hence be inferred that any mixture of oxygen and nitrogen will possess the average rate of effusion of its constituent gases .
4 .
Effusion of Air , Carbonic Oxide , Oxygen , and of a mixture of Carbonic Oxide and Oxygen at different pressures , by Plate B. The carbonic oxide was prepared according to Mr. FOWNES'S process , by heating oil of vitriol upon ferrocyanide of potassium : the gas was collected , as a measure of precaution , over alkali .
The arrangements were similar to the last ; barometer at 29-29 inches , thermometer 52 ? .
TABLF , III.-Effusion .
Air .
Carbonic oxide .
Oxygen .
Mixture of Gauge barometer in C 50 carbonic oxide + 50 oxygen .
inches .
I. I II .
II I. II .
I. , II .
28 O0000000 20 118 119 1 116 117 125 126 121 121 12 12Q0 0 11 1017 118 126 123 122 8 67 66 66 66 71 71 68 68 4| 81 82 80 81 86 86 83 2 83 2 .
556 56 55 55 60 60 58 58 442 443 435 436 468 469 453 452 The passage of the gases is somewhat quicker throughout than in the preceding experiments with the same plate , but the ratio between their velocities remains constant .
Comparing again the same portions of the scale , we have-Time in seconds .
Time , air 1 .
Gauge barometer .
I. .
i. Carbonic Oxygen .
xtre Carbonic Air .
Oxygenox i xdtOx e. Oxygen..Mixtre .
oxide .
oxide. .
Fron 28 to 12 inches ... ... ... 2385 234 251'5 43 0*9811 1-0545 1-0188 From 12 to 4 inches ... ... . .
148 146-5 157 151 0'9898 1-0608 1-0202 Taking 0'96779 as the specific gravity of carbonic oxide , the square root is 0*9838 , which corresponds closely with the observed time above , being intermediate between the times for the two different portions of the scale .
583 The time of effusion also of the mixture of carbonic oxide and oxygen in equal volumes is obviously the square root of the density of the mixture of the two gases : Observed time of mixture ... 1*0188 Calculated time of gases ... .
0182 The observed time of effusion of the mixture being within one thousandth part of the calculated time .
5 .
Effusion of Carbonic Acid , Air , and of mixtures of Carbonic Acid and Air , at different pressures , by Plate B. The arrangements continued the same as in last experiments ; barometer 29'58 in .
; thermometer 49 ? .
TABLE IV.-Effusion .
|Air .
Carbonic acid .
First mixture , i Second mixture , Third mixture , Gauge barom .
75CO2+25 air .
50C02+50 air .
25C02+75 air .
in inches ... ... ... ... ... ... ... ... . .
I. II .
I. II .
I I. I I. II .
I. II .
H/ /1fJ 28 0000000 00 00 20 121 121 145 146 141 140 135 ' 134 128 128 12 123 123 150 149 143 43 3 137 137 131 131 8 &gt; 69 70 83 84 81 80 : 76 77 74 73 4 85 84 103 103 98 99 95 94 90 90 90 2 58 59 71 70 7 68 67 64 64 61 61 456 457 552 552 531 529 507 506 484 483 ... ... , ... ... ... ... .
Comparing again the times ceding tables : in the two divisions of the scale adopted in the preTime in Seconds .
Carbonic acid .
Mixture I. Time of Effusion , time of Air = 1 .
Carbonic acid .
Mixture I. Mixture I. Mixture III .
Gauge barometer .
- ... ... . .
Observed .
Calculated .
Observed .
Calculated .
Observed .
Calculated .
Observed .
Calculated .
From28to 12inches .
12090 .
1-2365 1-1618 11818 1'1127 1 1245 1-0618 1'0647 From 12 to 4 inches .
1'2143 1 !
2365 The calculated number for carbonic acid ( 1'2365 ) is the theoretical time , or square root of the density of the gas ; the calculated times for the mixtures are also the square roots of the respective gravities of those mixtures .
The times of effusion of carbonic acid compared with air do not therefore differ 584 more than the numbeis 1V209 and 1P214 , in the two divisions of the scale ; or 1 part in 242 , a deviation which may be considered as within the errors of observation .
The mixtures of carbonic acid and air have also the mean times of the pure gases .
6 .
Effision of mixtures containing Hydrogen .
* [ I was induced to examine the effusion of mixtures of hydrogen and other gases very minutely , in order to elucidate if possible certain singular peculiarities which were observed in the transpiration of these mixtures by tubes .
A new plate E was employed , composed of thin platinum foil y-th of an inch in thickness , with a circular aperture -26th of an inch in diameter , as measured by Mr. POWELL by means of a micrometer .
It was desirable to simplify the experiment at the same time by operating upon a constant volume of gas , measured before effusion , and drawn into an aspirator-jar which was maintained vacuous , or as nearly so as possible , by uninterrupted exhaustion .
The gas was measured in a globular jar , to which more particular reference will be made hereafter .
It contained 65 cubic inches between two marks , one upon each of its tubular axes , and was supported vertically over the water of a pneumatic trough .
The little brass tube upon which the perforated plate is fixed ( a. in fig. 5 , Plate XXXIII .
) was now made to screw upon the end of one of the stop-cocks , namely , L ( fig. 1 .
) , which is immediately attached to the aspirator-jar , and projected upwards within the block tin tube H. The perforated plate was fixed to the end of its brass tube by means of soft solder .
The results thus obtained I consider superior in value to those already detailed , from the longer periods of observation , the time for air generally amounting to 800 or 900 seconds ; from the new plate being thinner and its aperture of a regular circular form ; and from the greater silnplicity of the conditions of the experiment , namely , the passage of the gases into a sustained vacuum under the whole atmospheric pressure .
The series of experiments is divided into five sections , each containing the experiments of one day , to which the height of the barometer and the temperature are added .
Two observations were made of the time of effusion in seconds for each gas , which are given under the columns of experiments I. and II .
, and the mean of the two experiments is added in a third column .
This mean is expressed in the column which follows , with reference to the time of oxygen as 1 .
The additional column headed " calculated times of mixtures , oxygen= 1 , " contains times of the mixtures , calculated from their specific gravities , being the square roots of the densities of the respective mixtures .
The observed times of the hydrogen mixtures will be seen to correspond very closely with these calculated numbers , the maximum divergences not exceeding that of pure hydrogen itself .
The principal results of the preceding table , and also the results of two series of experiments or mixtures of hydrogen with carburetted hydrogen ( C H2)and with carbonic acid , are exhibited by means of the curves projected in Plate XXXIV , for the pulrpose of comparing with them the results of the transpiration of the same mixtures exhibited in Plate XXXV , which I have not yet succeeded in reconciling with any physical law .
Feb. 1846 .
] The number-s at the top and bottom of the Plate , which apply to the vertical lines , express the times of effusion , the time of oxygen being taken as 100 ; while the numbers to the right of the table , and which apply to the horizontal lines , express the volumes of hydrogen in 100 volumes of the mixture .
Thus the curves all terminate above in a common point , 26*3 , the time of 100 hydrogen ; and each terminates below with the proper time of the particular gas which is mixed with hydrogen , the proportion of hydrogen being then 0 , and that of the other gas 100 ; that is , the curve of the carburetted hydrogen mixtures at 72*32 ; the curve of the nitrogen mixtures at 935 ; that of the air mixtures at 95'1 ; that of the oxygen mixtures at 100 , and that of the carbonic acid mixtures at 116 .
7 .
Effusion of r of different Elasticities or Densities , by brass plate B. In all the experiments hitherto described , the air or gas effused was under the atmospheric pressure , which varied only within narrow limits .
It was desirable to know whether the time remained constant for the passage into a vacuum of equal volumes of air of all densities , which the theory of the passage of fluids into a vacuum requires .
The air was drawn into the receiver of an air-pump ( fig. 2 .
Plate XXXIII .
) , maintained vacuous by continued pumping , from the globular gas receiver a , placed in a deep glass basin half-filled with water and used as a pneumatic trough ; this basin and the globular vessel being placed on the plate of a second air-pump under a large bell-jar in which a partial exhaustion could be maintained during the continuance of the experiment .
The vessel a had tubular openings at top and bottom ; its capacity between the marks b and c in these necks was 65 cubic inches ; the lower tube was expanded under the mark b into an open funnel ; the upper tube was cylindrical with a flange or lip , and had a sound cork fitted into it .
A short brass tube d , of quill size , soldered to the end of the stopcock e , descended into the bell-jar and passed through the cork of a , which was perforated .
The vessel a having thus an air-tight communication with the exhausted receiver v of the first air-pump , by the tube F , the drying tube U and the tube H ; a measured quantity of air ( 65 cubic inches ) could be drawn from it by observing the time which the water of the trough took to rise from the mark b to c. The perforated brass plate , through which the gas had to pass , was attached to the stopcock L , as before , and was therefore within the tube H. It is represented of one-fourth of its linear dimensions in fig. 5 , Plate XXXIII .
4 G2 587 When the large bell-jar over a was not exhausted , the gas in the latter was of the atmospheric tension .
With the barometer at 29'28 inches , and thermometer at 54 ?
, the air was withdrawn from the globe a , in 388 seconds in one experiment , and in 389 seconds in another .
The pressure upon the air in a was then reduced to three-fourths of an atmosphere , by exhausting so that the gauge barometer stood at 7'32 inches from the bottom of the scale , which is one-fourth of the whole pressure of 29*28 inches .
The globe a was thus occupied by air of the tension of three-fourths of an atmosphere , or 21'96 inches .
It was in this state connected with the vacuous receiver v of the airpump , and the time required for the effusion of the constant volume of 65 cubic inches of air , measured in its rarefied state , between the marks b and c , observed .
The effusion of this volume of air of three-fourths density was effected in two experiments in 389 and 392 seconds .
Again , the air in a being made of 14*64 inches tension , or half an atmosphere , the constant volume was effused into a vacuum in 411 and 408 seconds .
Lastly , with the air in a of 7'32 inches tension , or one-fourth of an atmosphere , the time of effusion was 438 and 439 seconds .
The results therefore of the effusion of a constant volume are as follows : Time of effusion .
Air of 1 atmosphere ... .
388*5 seconds ... 1 .
Air of 0'75 atmosphere ... .3905 seconds ... 10051 .
Air of 0'5 atmosphere ... 409'5 seconds. .
10541 .
Air of 0'25 atmosphere .
438'5 seconds. .
1*1287 .
It thus appears that the effusion of air into a vacuum is very little affected by a moderate change of density ; air of 1 atmosphere and of 0-75 atmosphere passing in nearly the same time .
The effect therefore of the ordinary changes of the barometer on the effusion of air must be small , if at all sensible .
A retardation occurs in the effusion of air of diminished density , which amnounts to an excess of -1-th of the time , on air of 0'75 tension ; of 2th on air of 0-5 tension , and -th on air of 0'25 tension .
Experiments were also made on the effusion of air of higher density than 1 atmosphere .
The air was drawn of any required tension from 1 to 2 atmospheres from a strong globular vessel A ( fig. 3 .
Plate XXXIII .
) , provided with a gauge barometer and mercury , by which the tension of the compressed air within it was observed .
Before its admission into this vessel the air was previously condensed in another vessel D by a syringe , to a higher degree of density than was required in A , and the supply of compressed air , regulated by the adjustment of an intermediate stopcock , so as to keep the gauge of A at a constant elevation , which could easily be done within -2oth of an inch .
In experiments with compressed air , the latter was allowed to flow into the twopint jar exhlausted on the plate of the air-pump , and the time observed which the gauge barometer required to fall through its range from 28 to 2 inches .
During the 588 following experiments the height of the barometer was 29*3 inches , which is the value of 1 atmosphere , and the thermometer 53 ? .
TIABLE VI.-Effusion of Air of different Densities .
Height of Air of 1 atmo Air of Air of Airof Air of Air of fgaugebarom .
sphere .
125 atmospatmosphere .
atmosphere .
1phere .
atmosphere .
2 atmospheres .
1 atmosphere .
in inches .
I. I. I. I. I. II .
I. II .
I I. II .
III .
IV .
t/ IJ t II II 1/ / / l / / 'II / / 28 00000000000 20 116 116 90 91 76 75 65 65 56 55 117 117 12 118 118 92 91 76 76 64 64 56 56 118 118 8 65 65 47 48 38 38 33 33 30 29 66 67 4 79 80 50 50 41 40 33 32 26 28 80 80 2 54 54 26 25 18 18 16 17 14 13 54 53 432 433 305 305 249 247 211 211 182 181 435 435 In these experiments the depression of the gauge barometer is not produced by a constant volume of the compressed air , but by a volume which is inversely proportional to the density of the compressed air ; half a volume of air of 2 atmospheres being equal in the aspirator-jar , on the plate of the air-pump , to a whole volume of air of 1 atmosphere .
Correcting the times of the preceding table , we have the passage of equal volumes of air of different densities , between the gauge height of 28 and 12 inches , as follows : Time of effusion of equal volumes .
Air of 1 atmosphere ... 234*5 seconds. .
1 Air of 1'25 atmosphere ... .
227*5 seconds. .
0*9701 Air of 1.5 atmosphere ... .
227*2 seconds. .
0*9688 Air of 1-75 atmosphere ... .
225'2 seconds. .
0'9603 Air of 2 atmosphere ' ... .223 seconds. .
0-9510 It appears then that air of different densities between 1 and 2 atmospheres is effused in nearly equal times , the time of effusion diminishing slightly , not more than 5 per cent. with air of double tension .
Taking the whole range of the preceding and present results , we have air varying in density from 0'25 to 2 atmospheres , or from 1 to 8 , while the extreme variation in the time of the effusion of equal volumes is from 0-9510 to 1'1287 , or from 1 to 1'1868 .
In the lower part of the scale a more sensible inequality is perceived .
Thus , with an exhaustion of from 8 to 4 inches in the aspirator-jar , the passage of equal volumes of air of different densities takes place in the following times : Time of effusion of equal volumes .
Air of 1 atmosphere ... .
145 seconds. .
1 Air of 1*25 atmosphere ... .
121'9 seconds. .
0 8407 Air of 1'5 atmosphere ... .
1177 seconds. .
0'8117 Air of 1-75 atmosphere ... .
114'6 seconds. .
0'7903 Air of 2 atmospheres ... .
113 seconds. .
07793 589 Here the time of effusion of air of 2 atmospheres fallsabout 22 per cent. below that of air of 1 atmosphere , while in the upper part of the scale the difference was only 5 per cent. [ 8 .
Effusion of Air of different Temperatures , by Plate F. This plate was a portion of thin platinum foil , with an aperture of an irregular hatchet form , of which the two greatest cross diameters were --1th and j:-oth of an inch .
The perforated plate was attached to the end of the little brass cylinder by means of soft solder .
A two-pint jar , giving a cavity of 72'54 cubic inches , was used as the aspirator-jar , and the time of the fall of the gauge barometer was observed from 28'5 to 23'5 inches , with the admission of dry air at different temperatures .
1 .
The temperature of the room being 41 ?
FAHR. , and the height of the barometer 29'616 inches , dry air entered the aspirator-jar in three experiments in 533 , 532 and 530 seconds , of which the mean is 531*66 seconds .
The room being afterwards heated up to 52 ?
, the time of effusion of an equal volume of air was found to be , in three experiments , 525 , 527 and 526 seconds , of which the mean is 526 seconds ; or , a rise of 11 ?
in temperature has shortened the time of effusion by 5*66 seconds .
Taking the density of air at 32 ?
as 1 , at 41 ?
it will be 0'9820 , of which the square root is 0'9909 ; and at 52 ?
it will be 0-9609 , of which the square root is 0'9802 .
Now the relative times of effusion observed , namely , 531'66 and 526 seconds , are as 0'9909 to 0 9803 , numbers which all but coincide with the square roots of the densities , 0'9909 and 0'9802 , at the two different temperatures .
2 .
With the barometer at 30'186 to 30'150 inches , experiments were again made on the effusion of the same volume of dry air at 380 , 48 ?
0 and 580 , four hours elapsing between each set of experiments , which were required to bring up the room and apparatus to a uniform and steady temperature .
In three experiments at each temperature , The time of effusion at 38 ?
was 526 , 527 and 526 seconds : mean 526'33 .
The time of effusion at 48 ?
was 520 , 521 and 520 seconds : mean 520'33 .
The time of effusion at 58 ?
was 515 , 516 and 515 seconds : mean 515'33 .
Here the first rise of 100 shorttens the time of effusion 6 seconds , and the second rise of 10 ?
shortens the time 5 seconds more .
The density of dry air being 1 at 32 ?
, it is at 380 , 09879 ; at 480 , 0'9684 $ and at 580 , 09497 , of which three last densities the square roots are 0'9939 , 0'9841 and 0'9745 respectively .
Now the three mean times of effusion observed are in the proportion of the numbers 0'9939 , 0'9826 and 0'9731 , which correspond more closely with the preceding square roots than could be expected from the nature of the experiments .
It appears then that the effision time of air of dtffrrent temperatures is proportional to the square root of its density at each temperature .
The velocity of the effusion will be inversely as the square root of the air 's density .
Hence two volumes of air which have not the same temperature , are , in regard to effusion , like different gases possessing the densities of the air at the two temperatures .
As the velocity of the effusion of air does not increase at a rate so rapid as the direct proportion of its expansion by heat , it follows that the flow of air under pressure , through a small aperture , is retarded by heating the air ; that is , the same absolute quantity or weight of air will take a longer time to pass , when rarefied by heat , than when in a dense state .
I have made several experiments on the influence of aqueous vapour upon the effusion of air .
When dry air was effused into an aspirator-jar with the gauge barometer attached , and immediately afterwards air saturated with moisture at the same temperature , the latter passed through in sensibly the same time with comparatively large apertures , but in a shorter time with small apertures , although in general without much uniformity in successive experiments .
Thus the time for dry air being constant at 524 seconds with plate F of small aperture , barometer 29'812 , and thermometer 49 ?
; with moist air , the time gradually fell , till at last it appeared to settle at 506 seconds , that number being obtained in three successive experiments ; the temperature in the mean time having risen to 51 ? .
There is here an acceleration of 18 seconds , of which not more than 2 seconds are accounted for by the diminished density of the moist air , and 1 second more by the rise in temperature .
The moist air seemed also to have an extraordinary effect in opening and enlarging fissures , and very soon rendered more than one platinum plate useless , which was fixed by brazing , by that action .
Nov. 1847 .
] PART II .
TRANSPIRATION OF GASES .
1 .
Transpiration of Air of different Densities or Elasticities , by a Glass Capillary T'ube E. ( a. ) The same arrangements were adopted as in the effusion of air of different densities , lately described , the capillary tube being interposed in the place of the perforated plate .
The apparatus employed is represented in fig. 4 .
Plate XXXIII .
With barometer 29*28 , and thermometer 54 ?
, 65 cubic inches of dry air of the atmospheric density were transpired from the globular vessel a ( fig. 2 ) , into a good vacuum sustained by continued pumping , through a capillary glass tube E , twenty feet in length ; the same volume of air of 0'75 atmosphere and 0'5 atmosphere , measured at these pressures , were also transpired by the same capillary .
The times were as follows : I. II .
Mean .
Iv I. Transpiration of air of 1 atmosphere ... ... ... 799 800 799'5 Transpiration of air of 0-75 atmosphere ... ... ... 1049 1051 1050 Transpiration of air of 0'5 atmosphere ... ... ... 1545 1542 1543'5 591 It is obvious that the times approach the inverse ratio of the tensions , as will appear more clearly on comparing the times observed with those calculated on that principle .
Time observed .
Time calculated .
Transpiration of air of 1 atmosphere ... ... ... 1 Transpiration of air of 0'75 atmosphere ... ... ... 1-3133 1-3333 Transpiration of air of 0'5 atmosphere ... ... ... 19306 2 With air of higher tension than 1 atmosphere , the same apparatus for compression was also employed as in the effusion experiments ( fig , 3 ) .
The capillary E communicated with the two-pint aspirator-jar ( capacity 72-54 cubic inches ) , which was fully exhausted on the plate of the air-pump .
The air being then allowed to pass into the capillary , the instant of time was noted when the gauge barometer fell to 28 inches , and the other points described below .
The external barometer stood at 29'08 inches ; thermometer at 53 ? .
TABLE VII.-Transpiration of Air of different Densities .
Air of Air of Air of Air of Air of Air of Gauge barometer 1 atmosphere .
2 atmospheres .
1'75 atmosphere .
1'5 atmosphere .
1'25 atmosphere .
1 atmosphere.in inches .
I. | .
I I. I II .
I. II .
I. II .
I. II .
III .
l/ , , / t l , , , , .
/ / / / / / 28 00000000000 20 253 254 66 67 85 86 114 114 162 161 255 12 310 310 70 69 91 90 123 124 182 179 311 8 220 220 37 37 49 49 69 70 110 109 221 4 349 352 39 40 53 53 78 77 132 134 346 2 328 328 20 19 29 28 41 41 78 79 229 1460 1464 232 232 305 306 425 426 664 662 1462 The times of transpiration in the preceding table require to be corrected , as they iepresent the passage of equal volumes of air measured after and not before the transpiration .
Thus the same depression of the gauge barometer would be produced by half a volume of air of 2 atmospheres , as by a whole volume of air of 1 atmosphere ; and it is necessary therefore to double the times observed of air of the former density , to obtain the time of passage of a whole volume .
For the transpiration of equal volumes , in the gauge-range from 28 to 20 inches , which is most nearly equivalent to the action of a vacuum , we haveEqual volumes .
Observed time of transpiration .
Calculated time .
Air of 1 atmosphere ... ... ... 254 seconds ... ... ... 11 Air of 1'25 atmosphere ... ... ... 201*9 seconds ... ... ... 0-7949 0'8000 Air of 1*5 atmosphere ... ... ... 171 seconds ... ... ... 0-6732 0'6666 Air of 1'75 atmosphere ... ... . .
149*6 seconds ... ... ... 05890 0'5714 Air of 2 atmospheres ... ... 133 seconds ... ... . .
0*5236 0'5 .
_ I,1 1 S. I 092 The calculated times of the last column are the reciprocals of the tension , or number of atmospheres in the first column ; they represent the observed times within a sufficient degree of approximation , to prove thatfor equal volumes of air of difffrent densities , the times of transpiration are inversely as the densities .
The velocity of transpiration will therefore be directly in proportion to the density of the air ; air of double density being transpired into a vacuum in half time .
This at once separates the action of a capillary tube from that of a minute aperture ; for air of all densities , it will be remnembered , passes into a vacuum by effusion with equal velocity .
A consequence of this law immediately appears in conducting transpiiation experiments , in the marked influence of the height of the barometer on the time of transpiration ; the higher the barorneter and the denser the air , the more quickly does a constant volume of it pass through a capillary tube into a vacuum .
This appears also to separate transpiration from the ordinary action of friction , for the denser the air , the more should its passage be retarded by friction .
[ 2 .
Transpiration of Air of different Temperatures .
Dry air was transpired by a glass capillary tube K , of fine bore , 39'4 inches in length , into a two-pint jar till the gauge barometer fell from 28'5 to 23'5 inches , in 796 , 794 and 794 seconds , in three successive experiments , made at the temperature of 41 ?
FAHR. , and with the barometer at 30'052 inches .
Four hours afterwards , the air and all the apparatus having been for some time at 58 ?
, an equal volume of dry air was transpired twice in 814 seconds .
A difference of 17 degrees of temperature has made a difference of 19 seconds in the time of transpiration , and the dense cold air is transpired most rapidly .
The times are nearly in the inverse ratio of the square root of the densities of air at the two temperatures .
The transpiration of air in the first experiments which are made in the morning is often observed to be more rapid than in those which follow , owing I believe to the low nocturnal temperature being retained for some time by the glass capillary .
January , 1847 .
] 3 .
Preliminary Experiments on the Transpiration of dferrent Gases by Capillary Tubes , A , B and C. The times of transpiration of the gases will be expressed in the sequel with reference to the time of oxygen as unity instead of that of air .
Assuming what is now almost universally conceded , that the atomic weights of the following elements are exactly expressed by entire numbers , namely , oxygen by 8 , nitrogen by 14 , carbon by 6 , and hydrogen by 1 , and that while the equivalent proportion of the first affolds one volume of gas , that of each of the others affords two volumes , we obtain the following theoretical densities for these.elements and several of their gaseous compounds .
The experimental determinations which appear to be of most value are subjoined .
of Specific Gravities of Gases .
Air =1 .
Oxygen =1 and 16 .
Calculated .
Observed .
Calculated .
Oxygen ... ... ... ... ... ... Nitrogen ... ... ... ... ... ... Air ... ... ... ... ... ... ... ... Hydrogen ... ... ... ... ... Carbon ... ... ... ... ... . .
Carbonic acid ... ... ... . .
Nitrous oxide ... ... ... ... Nitric oxide ... ... ... ... Carbonic oxide ... ... ... Carburetted hydrogen Olefiant gas ... ... ... ... Sulphuretted hydrogen 1.1099* 0-9712 1 0-06937 0'4162 1-5261 1-5261 1-0405 0-9712 0-5549 0-9712 1'1793 1*10563 REGNAULT .
1 0-97137 1EGNAULT .
0-8750 1 0'9010 0-06926 REGNAULT .
0-0625 ... ... ... ... ... ... ... ... ... ... ... ... 0'3750 1-52901 REGNAULT .
1-3750 ... ... ... ... ... ... ... ... ... ... ... ... 1-3750 ... ... ... ... ... ... ... ... ... ... ... . .
0 ... ... ... ... ... ... .0-9375 0-9678 WREDE .
0-8750 0 ?
555 { 0IIOf SON and 0 5000 ' IIENRY .
J 0-9852 SAUSSURE .
0-8750 1.1912 { GAY-LUSSAC 1 0625 and TIIlENARD 16 14 14-416 16 22 22 15 14 8 14 17 1 0-8785 REGNAULT .
0-9038 REGNAULT .
0*0626 REGNAULT .
1*3830 REGNAULT .
0-8754 WIEDE .
0-5001 1 IIOMSoN and 0 HENRY .
0-8904 SAUSSURF .
1'0766 GAY-LusSAc and ITHENARD .
With the exception of the recent valuable determinations of M. REGNAULT and Baron WREDE , the carlculated specific gravities are probably nearer the truth and more to be depended on than the experimental results found in books , which arc old , and generally not made with that degree of precision which the science now requires .
Capillary A.-This glass tube was thiity inches in length , and of a fine cylindrical bore ; it allowed 1 cubic inch of air of the usual tension , to pass into a vacuum in about 13 seconds .
A pint-jar was exhausted , of which the capacity , including the vacuous spaces of the air-pump , was 41V64 cubic inches .
The gas entered into this space , passing through the capillary , and depressed the attached gauge barometer in the times stated in the following Tables : TABLE VIII.-Transpiration by Capillary A into a one-pint jar .
Barom. 29'55 .
Temp. 61 ? .
Air .
Oxygen .
Hydrogein .
Gauge barometer in inches .
IL .
I. I. II .
I. II .
, , ... ... . .
, , 28 00000 20 150-5 150 166*5 166 71 70'5 12 181.5 182 201 201.5 87 87 4 321 321 354*5 354 156 155 2 ... ... ..3 ... ... ... ... ... .
2135 215'5 94-5 94-5 From 28 to 4 inches ... ... .
653 653 From 28 to 2 inches ... ... ... ... ... ... ... . .
9355 937 408.5 407 It is a question how far this variation in the ratio is owing to the action of effusion ; the time of effusion of this gas being only 0'25 , referred to oxygen as unity , while its time of transpiration is 0'4355 .
The influence of effusion upon the rate of passage is likely to be most considerable when the pressure is greatest ; and that of transpiration , on the contr ary , most considerable when the resistance to the passage of the gas is greatest and the pressure least , that is , in the lower part of the scale .
It may be observed that the transpiration time of hydrogen does not differ far from 0'4375 , which is one-half of the transpiration time of nitrogen , calculated from the experiment on air , or seven-sixteenths of that of oxygen .
Carbonic acid appears to be much more quickly transpired than oxygen , although denser than that gas in the ratio of 11 to 8 ; but the effusion time of carbonic acid being slow , any influence of effusion will increase the time of transpiration of this gas , the reverse of what occurs with hydrogen .
The transpiration time of carbonic acid varies considerably at different pressures , being slower by 0*0535 at the upper than the lower part of the scale .
An approach to 0175 , or twelve-sixteenths of the time of oxygen , may be noted at present in the rate of this gas .
The transpiration rate of carburetted hydrogen appears to be affected by an error of observation in the middle part of the scale ; but its rate is slower at the upper part than at the lower , to the extent of 0'0211 .
This is also in accordance with the assumed influence of effusion , the effusion time of this gas being greater than its transpiration time .
The transpiration time of carburetted hydrogen is not in direct proportion to its gravity , which is 0-5 , or one-half of that of oxygen : it approaches more nearly to 0'5625 , which is nine-sixteenths of the time of oxygen .
Capillary B.--This glass tube was 31-5 inches in length , of a round bore , but decidedly conical .
It was first placed so that the gas entered the tube by the large and escaped by the small opening .
When so arranged this tube allowed 1 cubic inch of air to pass into a vacuum in 34'3 seconds , or the transpiration was nearly three times slower than by A. TABLE X.-Transpiration by Capillary B into a half-pint jar ( 21'26 cubic inches ) .
Barom. 29'17 .
Temp. 68 ? .
Gauge barometer in inches .
xygen .
Nitrogen .
Carburetted hydrogen .
Gauge barometer in inches .
I. II .
I. II .
III .
I. II .
/ / / I , I / / / / I/ I , 28 0000000 20 235-5 235-5 209 205 205 133 133 12 289 290'5 252 253 ... ... 162'5 163 4 521 521 451-5 453-5 ... ... 290'5 290'5 2 290'5 290 258 255'5 ... .
164 165-5 Fr~om 28 to 2 inches ... 1336 1337 1170-5 1167 '205 70 752 From 28 to 2 inches ... 1336 1337 1170'5 1167 205 750 752 596 Mean Results .
Carburetted Gauge barometer .
Nitrogen .
rre hydrogen .
From 28 to 20 inches r Time in seconds 2 ... ... 06'3 133 Time of oxygen-= ... 0'8760 0-5647 Time in seconds ... ... .
252'5 162-7 From 20 to 12 inches{ Time in Aseconds 252'S , 6'7 From 0 to 1 inches Time of oxygen= 1 ... 0-8716 0-5618 From 12 to 4 inches Time in seconds 452*5 290'5 From 12 to 4 inches c. pTime of oxygen 1= ... 0'8685 0*5576 rom 4 to 2 inches ... Time in seconds ... ... 256'7 164-7 *Time of oxygen= ... 0-8845 0'5677 From 28 to 2 inches |Time ill seconds ... ... 1168'7 751 Time of oxygen=l ... 0'8740 0'5619 TABLE XI.--Transpiration by Capillary B into a half-pint jar .
Barom. 29'2 .
Oxygen .
Hydrogen .
Carbonic acid .
Gauge barometer in inches .
- .
, I. II III .
I. II .
III. .
I II .
III .
62 ? ?
62 ?
-5 64 ?
*5 65 ? ?
65 ? ?
65 ?
.5 66 ? ?
66 ? ?
66 ? ?
28 0 " 0/ 00 " 0 " 0 ' 0 ' 0 0 " 20 234 234 234-5 ]03 103 103 174'5 174 173 12 285 287-5 288*5 126i5 126 126*5 2]2 211'5 212 4 516 516 516*5 229-5 228 228 376 379-5 376 2 290 292-5 288-5 129 126 132 ... ... ... 213 212'5 620.5 64 ? ?
64 ?
'5 65 ? ?
650'5 ... ... ... ... ... 66 From 28 to 2 inches ... 1325 1330 1328 588 583 589,5 ... ... ... 978 974 Mean Results .
Gauge barometer .
Hydrogen .
Carbonic acid .
/ ioQ m on f.Time in seconds ... .
103 173'8 From 28 to 20 inches Time of oxygen. .
04398 0'7424 rom 2o .
os ihe{ Time in seconds ... ... ... 126-3 211*8 rom 20 to 12 nche Time of oxygen= 1 ... . .
0'4403 0'7385 -r To 4 .
A carbonic acid into a perfect vacuum would certainly approach still more nearly to 0'75 ; for the rate of transpiration of that gas appears , in the case of the present capillary , to become slower with the increase of pressure .
It will be seen hereafter that the gases deviate very sensibly at low pressures from the empirical coefficients of transpiration which have been named , becoming slower in their passage with reference to oxygen , as is here observed of carbonic acid .
This deviation appears to be connected with excessive resistance , whether arising from the smallness of the capillary opening or diminished pressure .
With the view of observing the effect of alterations in the position and dimensions of a capillary , experiments were now made with this tube , ( 1 ) in an inverted position , so that the gas entered by the narrow and escaped by the wide end , and ( 2 ) after being reduced to half its original length .
Capillary B reversed.-The passage of gas through this tube was nearly three times more rapid in the new direction , 1 cubic inch of air being now transpired into a vacuum in 12'6 seconds .
The length of the tube and resistance to passage through it are therefore nearly the same as in capillary A ; but while the latter was of uniform bore , the present capillary is highly conical .
TABLE XII.-Transpiration byCapillaryB reversed into a one-pintjar .
Barom.30'13 .
Air .
Oxygen .
Hydrogen .
Carbonic acid .
Gauge barometer in inches .
I. II .
III .
I. I.L III .
I. .
I. I. I. IIE .
59 ?
59 9 ? .
59 ?
F. 61 ?
F. 61 ?
-5F .
61 ?
'5F .
62 ?
F. 62 ?
F. 28 0 ' 0 ' 0 ' 0 ' 0 ' " 0pt 0 " 0 ' " 0 ' 0't 0 ' 20 149-5 144-5 145 160 160 160 69 69 127 126 126 12 173-5 177-5 178 198 196 196 86 85-5 154 153 152 4 318 318 317 352'5 353*5 353*5 157-5 157-5 268'5 267-5 268*5 2 190 189*5 192'5 216 216-5 215 96 96'5 154'5 155 156'5 From 28 to 12 inches ... 323 322 323 358 356 356 155 154'5 281 279 278 From 28 to 2 inches ... 831 829-5 832-5 926-5 926 924'5 408-5 408-5 704 701-5 703 Mean Results .
Gauge barometer .
Air .
Hydrogen .
Carbonic acid .
From to12 { inchesTime in seconds ... ... 322 154'75 279-33 From 28 to 12 inches Time of oxygen= 1 ... 0'9024 0-4338 0-7831 From 142to 4 inchesf Time in seconds ... ... 317-7 157'5 268-16 From 12 to 4 nches Time of oxygen= 1 ... 08994 0-4459 0-7595 From 4 to inches Time in seconds ... ... 190-7 96-25 155-33 From 4 to 2 incheTime of oxygen= 1. .
0-8836 0'4459 0-7197 From 28 to 2 inches Time in seconds ... ... 831 408-5 702-83 i Time of oxygen 0-8978 0'4413 0'7594 The coefficients of air and hydrogen , although still corresponding very closely with the numbers 0'9038 and 0'4375 , begin to exhibit a sensible variation in different parts of the scale .
The variation in carbonic acid is considerable , amounting to 0'0634 , and the divergence is on both sides of the empirical number 0'75 .
698 Capillary B shortened.-The tube was preserved in its last position , but its length reduced to 14'5 inches .
It now allowed 1 cubic inch of air to pass into a vacuum in 6'4 seconds ; or twice as rapidly as when entire .
TABLE XIII.-Transpiration by Capillary B ( 141 inches long ) into a one-pint jar .
Barom. 30'12 .
Air .
Oxygen .
Nitrogen .
Carburetted hydrogen .
Hydrogen .
Carbonic acid .
Gauge barometer in inches .
Iin inches. .
IIII .
II .
I. I. III ... .
III .
I. II II .
I. II .
620 F. 62 ?
F. 62 ?
F. 630 F. 630 F. 62 ?
F. 61 ?
F. 62 ?
F. 62 ?
F. 62 ?
F. 620 F. 62 ?
F. 620 F. : as28 O 0t ' 0 " 0o 0 " 0 " 0 " 0 " 0 " " 0 0 " 0t " 00 " 0 " 0 " 20 75 75 75-5 82'5 83-5 82 72 72 48 48 48 345,34 34 69 69 12 89-5 89-5 89-5 99'5 99-5 100 875 88 57 57 57 41.5 415 41-5 81 80-5 4 158 157 157-5 175 175-5 174-5 152-5 152-5 98 98 98 75 75 75 137-5 136-5 2 91-5 91 92 102 103 102-5 91 89'5 56 56 55-5 45-5 45-5 45 77 77-5 nw fi. .
-_ __ ____ ___ ___ ___ ___ From28tol12in .
164-5 164-5 164-5 182 183 182 159-5 160 105 105 105 76 75-5 75-5 150 149-5 From28to 2in .
414 412-5 414-5 459 461-5 459 403 402 259 259 258-5 196-5 196 19955 '364 363-5 Mean Results .
Gauge barometer .
Air .
Nitrogen .
CarburettedHydrogen .
Carbonic acid .
hydrogen .
From 28 to 12 inches f Time in seconds ... ... 164-5 15975 105 75-66 1497 Time of oxygen=1 ... 0-9023 0-8763 0-5728 04150 0-8211 om 12 to 4 ih Tinle in seconds ... .
157-5 152'5 98 75 137 From 12 to 4 hes me of oxygen = ... 09000 08714 0-5600 0-4285 0'7828 From 4 to inch f Time in seconds ... ... 91-5 9025 55-83 45-33 77.25 From 4 to 2 inches Time of oxygen=l ... 0-8926 0-8804 0-5447 0-4325 0-7546 From P8 t2 hince { Time in seconds ... ... 413'7 4025 258-83 196 36375 From 8to 2 inches Time of oxygen= I ... 0-8999 0-8755 0-5632 04263 07912 The times are now too short for accurate numerical determinations , but it is obvious that while the relative times of air and nitrogen are little changed , particularly from 28 to 4 inches , the times of hydrogen and carbonic acid are sensibly affected by effusion , and most so in the upper part of the scale ; the coefficient of hydrogen falling to 0'4150 , while that of carbonic acid rises to 0'8211 .
From 4 to 2 inches , the rates are 0'4325 and 0'7546 ; numbers which still diverge a little from the empirical coefficients , but both in the direction of the effusion influence .
Reduced to 7 inches in length and now allowing 1 cubic inch of air to pass into a vacuum in 3'4 seconds , this capillary was found to be still less adapted for transpiration .
In a series of observations , which are not of sufficient importanme to be particularly detailed , the coefficients of the gases in the range from 28 to 12 inches , were , -air 0'9194 , nitrogen 0'8983 , carburetted hydrogen 0'6029 , carbonic acid 0'9028 , and hydrogen 0-3930 : numbers which demonstrate an increasing interference of effusion .
In the range from 12 to 4 inches , the coeffcients were , -air 0'9114 , nitrogen 0'8851 , carburetted hydrogen 0'5764 , carbonic acid 0-8303 , hydrogen 0'4180 .
Of this last portion of capillary B , 5-5 inches were found to contain 2*78 grains of mercury ; which gives the tube a mean diameter of 0O0137 , or A--rd of an inch .
Capillary C.-This was a tube of exceedingly fine bore ; 8-3 inches of the tube 599 containing only 0'65 grain of mercury ; which gives a diameter of 0*00539 inch , or about , soth of an inch .
Experiments were made with portions of this tube of different leng , ths ; and first with a portion only 1 inch in length , in which it was expected that the influence of effusion would be considerable , from its approach to an aperture in a thin plate .
TABLE XIV.-Transpiration by Capillary C ( 1 inch long ) into a one-pint jar .
Barom. 28*81 .
Temp. 60 ? .
Air .
Oxygen .
Hydrogen .
Carbonic acid .
Gauge barometer in inches .
I. II .
I. II .
I. II .
I7/ fI t ; I t 28 0000000 20 248-5 249 273 2 ~71-5 108 ' 108 237'5 12 283 281'5 308-5 3085 126 127 261 8 1865 185 ' 5 2045 206 87 87 167 4 284 1 286 315-5 314 136 137 244 2i 270'5 266*5 301 302 133 ' 132 227 From 28 to 2 inches ... . .
1272'5 1268'5 1402-5 1402 590 591 1136-5 !
t : ''W '1 ''I W The velocity with which air passed into a vacuum was , by CI inch , 1 cubic inch in 21'26 seconds ; by C2 inches , 1 cubic inch in 35'43 seconds ; and by C4 inches , 1 cubic inch in 70 seconds .
C1 inch exhibits ( Table XIV .
) great variation in its rate of transpiration at different pressres , as was to be expected .
At the head of the scale , the rate of air is 0'9138 , of hydrogen 0'3967 , and of carbonic acid 0'8725 ; which all indicate a great interference of effusion .
At the bottom of the scale , on the other hand , where the pressure is small and the resistance to passage consequently great , the rate of air is 0'8905 , of hydrogen 0'4374 , and of carbonic acid 0'7529 , or nearly normal .
With C2 inches ( Table XV .
) , the rates of air are pretty uniform at different parts of the scale , and sufficiently norual ; the deviation from 8 to 4 inches appears to be an accidental anomaly .
The rate of hydrogen also is never distant from 0'4375 , except at the very bottom of the scale .
This appears to be the length of a tube of so small a bore which gives the most uniform results at different pressures : For with C4 inches ( Table XVI .
) , where the resistance is excessive , air has a rate corresponding sufficiently with its specific gravity at the head of the scale , but diverging rapidly in the lower part of the scale , the time of transpiration becoming rapid as compared to that of oxygen .
With a tube then like the present , the relation between the times of transpiration is only to be looked for witthin a limited range , and that at a high degree of pressure approaching to a whole atmosphere .
4 .
Trampi rtion by Capillary Tube H. preceding experiments on transpiration render probable , would be a point of fundamental importance in their history , I was induced to try new capillary tubes and to multiply experiments , varying the circumstances in which they were made , and taking additional precautions against the interference of disturbing causes .
rlThe arbitrary nature of the coefficients of transpiration indeed produces a more than usual necessity for strong evidence and numerous confirmations , as the numbers themselves have no a priori probability in their favour .
It appeared desirable to try capillary tubes of larger diameter than those already employed , and of great length ; partly to vary the conditions of the experiment , and partly because extreme shqrtness of tube appears , in the experiments made with C , to be unfavourable to uniformity of rate at different pressures .
Several portions of capillary tube , of which the bore was as nearly equal as coutld be judged of by the eye , were accordingly selected and cemented together at the blowpipe , so as to form a continuous tube 22 feet in length , which was bent up into coils for convenience in using it , as represented in figure 4 , Plate XXXIII .
The extremities of this capillary were connected with the block-tin tubes proceeding from the drying tube and air-pump jar respectively , by means of thick caoutchouc adopters , which diminished the rigidity of the arrangement and protected the glass tube from the effect of the shocks , which are unavoidable in working the airpump .
This capillary ( H ) allowed 1 cubic inch of air to pass into a vacuum in 15'64 seconds .
Two inches of the tube were found to hold 2*65 grains of mercury , which gives a diameter of 0'0222 inch , or -th of an inch .
The tube , however , may have been more contracted at the bendings .
Air-pump jars of greater size were also employed , so as to protract the time of passage , and give larger numbers .
Of these vessels , which I have termed aspirator jars , the capacity of the " three-pint jar " was 103'56 cubic inches , and that of the " six-pint jar " 201 78 cubic inches ; the vacuous channels of the air-pump and connecting tubes being included in these measurerments .
The temperatures recorded are those of the interior of the aspirator jar , observed )by a very small mercurial thermometer , containing no more than 50 or 60 grains of mercury , and therefore highly sensitive .
It will be remarked that the temperature within the receiver always rises , and in general about half of a degree FAHRENHEIT during the continuance of an experiment : this is owing to the compression of the gas already in the receiver by that which enters .
The change of temperature due to this cause becormes less considerable with large aspirator jars , which atre preferable to small jars for this and other reasons .
As it appears that the relation between the coefficients of transpiration is only to be looked for at high pressures , attention sihould be more directed to the rates of transpiration in the upper than in the lower part of the scale .
The inquiry will therefore be directed with the view of solving the question , What are the coefficients of transpiration of the different gases into a vacuum , or under considerable pressures ?
TABLE XVII .
L-Transpiration by Capillary I. into a three-pint jar .
Baroin .
30'324 .
Air .
Oxygen .
Hydrogen .
Hyd .
95+5 air .
Carbonic acid .
Gauge barometer in inches .
L. 1 .
II .
I. I. II .
I. II II .
I. II .
6105 61 61 ?
-5 15 6105 610.5 1 610*5 61 ?
5 61 ?
5 6105 610.5 28 !
I 0 " 0 ' 0 0 ' 0 ' 0 ' 0 " 0 " Q0 24 213 213 236 37 104 104 125 124 i 178 177 20 227 227 254 i 53 111 111 132 13 12 188 188 12 549 549 611 611 268 269 321 321 454 455 8 389 388 435 435 192 192 227 227 321 320 61i.75 61Ic*75 61 ?
i75 610.75 61-75 61075 61c.75 6'1'75 6 175 61 ?
75 Fron 28 to 8 inches ... . .
1378 1377 1536 1536 675 676 805 804 1141 1140 Mean Results .
Ga.uge barometer .
: Air .
Ilydrogen .
95 Hyd.+5 air .
Carbonic acid .
From2 to 24 inches f Time in seconds ... ... 13 104 1245 177s* " Time of oxygen i i ... : 0-9006 0-4397 0'5264 0'7505 From 24 to 0 in chesJ Time in seconds ... 227 1111 132 188 Time of oxygen= I ... 08954 0'4378 0'5206 0*7416 Fron 20 to 12 inhesf T ime in seconds ... ... 549 268*5 321 454'5 inches Time of oxygen= 1 ... 0'8985 0'4394 0'5252 017438 Frf 12 to 8 The numbers for air , hydrogen and carbonic acid accord well with those obtained by the other capillaries .
The small addition of 5 per cent. of air to hydrogen has a surprising effect in retarding the transpiration of that gas .
The mean rate of such a mixture , calculated from the rates of air and hydrogen separately , is 0*4625 , whereas the actual rate is 0'5264 .
The rate of the mixture should only be increased by 0'0228 , whereas it is really increased by 0'0840 .
Hence the effect of 5 per cent. of air in retarding the rate of hydrogen is nearly four times greater than it should be by calculation .
The experiment shows the effect which a small amount of impurity must have in deranging the transpiration rate of that gas .
I shall return again to this point under the subject of the transpiration of mixed gases .
TABLE XVIII.-Transpiration by Capillary H into a three-pint jar .
Barorn .
30'3-30'242 .
Mean Results .
In the preceding table , a comparison is made between two gases , nitrogen and carbonic oxide , of which the theoretical specific gravities are the same , namely 0'8750 , while the mean rate of nitrogen , from 28 to 12 inches , proves to be 0'8707 , and that of carbonic oxide 0'8678 , which is a pretty close approximation .
In several other experiments with these gases , a slight difference in their coefficients of transpiration was observed , carbonic oxide having always the smaller number : the difference generally approached 0'0040 .
It is not at all impossible , however , that these gases Oxygen .
Nitrogen .
Carbonic oxide .
Air .
Gauge barometer in inches .
I.I. I.I I. I.II .
I. II .
61 ?
.5 61 ?
-5 61 ?
-5 61 ?
-5 610-5 610.5 61 ?
5 61C.5 28 0 0 ' 01 0 ' 0 0 " 0 0 ' 24 238 239 208 208 207 20 214 214 20 253 253 220 221 220 220 229 229 12 614 614 536 533 532 533 552 550 8 434 434 378 376 376 377 389 392 610 ?
75 61 ?
-75 62 ? ?
62 ? ?
62 ? ?
620 620 62 ?
From 28 to 8 inches ... ... 1539 1540 1342 1339 1335 1337 1384 1385 Gauge barometer .
Nitrogen .
Carbonic oxide .
Air. .
Time in seconds ... . .
208 207 214 From 28 to 24 inches Tie of oxygen=l ... 018721 0-8679 0-8972 From 24 to 20 inhes Time in seconds ... ... 2205 220 229 From 24 to 20nches Time of oxygen= ... 018715 08695 0-9051 From 20to incs Time in seconds ... ... 534'5 532-5 551 From 20 to 12 inches Time of oxygen 1 ... 0-8705 0'8673 0-8974 From to [ ihes Time in seconds ... ... 377-5 376-5 390-5 From 2to 8cesTime of oxygen=l ... 0-8698 0-8675 08998 { Time in seconds ... ... 13405 1336 1384-5 From 28 to 8 inches Time of oxygen= ... 0-8707 0-8678 0-8993 604 Tables XIX .
and XX .
exhibit the transpiration rates of carburetted hydrogen and hydrogen , the former of which approaches to 0'55 , which is certainly a sensible deviation from 0'5625 .
The number for hydrogen ( 0'4418 ) , on the other hand , is but very little removed from 0'4375 .
TABLE XXI.-Transpiration by Capillary H into a six-pint jar .
Barom. 30 158-30 138 .
Oxygen .
Carbonic oxide .
Nitrogen .
Nitrous oxide .
Carbnic acid .
Gauge barometer in inches * ... . .
I I. 11 .
I. | II .
| II. .
1 .
II .
I. II .
I| 1 .
IL 67 67 670 67 670 67075 670-756 680 6s80 6s 680 28*5 o0 ' 0 ' 0 0 ' ooo o0 ' 00o 0"[ 26'5 230 231 20Q2 00 201 202 t2 1 173 173 173 173 24-5 237 235 202 203 204 205 205 176 175 174 176 23*5 111 1071 121 7 101 108 91 91 92 92 167 ' 670 67 ?
" 625 67C25 67c625 8 68 ? ?
68 ? ?
8 ? ?
68 ? ?
8 68 ?
Frofnt28*6 to 23'5 inches 588 587 55 11 515 515 440 439 439 441 _ __~ ?
j s _10 !
08 _ _1 , 91 .
' , 9i9 __ Mean Results .
Gauge barometer .
Carbonic oxide .
Nitrogen .
Nitrous oxide .
Carbonic acid .
From 285 to 239 inehe Time in seconds ... .
5105 515 i 439'5 440 Time of oxygen= I ... ' 0-8689 0 8766 0-7480 0-7455 In the preceding table carbonic oxide and( nitrogen are again compared , and also another remarkable pair of gases having the same theoretical density , namely carbonic acid and nitrous oxide .
The two latter exhibit an extraordinary parallelism in their rates of transpiration in all experiments which were made upon them , provided due attention was paid to the purity of the nitrous oxide .
The solution of nitrate of ammonia should always be filtered , and the salt crystallized , as the presence of a very minute quantity of solid matter may cause a change in the mode of decomposition of the salt by heat , and the evolution of a very sensible quantity of free nitrogen .
The carbonic oxide of experiments 1 and 2 was obtained by the action of oil of vitriol on pure oxalic acid ; that of experiment 3 was prepared by the process of Mr. FOWNES , namely heating oil of vitriol upon the ferrocyanide of potassium , avoiding a violent reaction by a proper regulation of the temperature .
The gas of both processes was washed with alkali , although this precaution is scarcely required with the gas of the last process .
The transpiration results are exactly the same with the gas prepared in both ways .
606 XXII.-Transpiration by Capillary H ( with cupped ends ) into a six-pint jar .
Baroni .
29'39 .
Temp. 51 ? .
Nitrogen .
Oxygen .
Olefiant gas .
Air .
Carburetted hydrogen .
Gauge barometer in inches .
I II .
I. II I. II I II I II .
III .
IV .
28-5 00000000000 26-5 203 204 233 235 122 121 211 211 129 129 129 129 24-5 206 207 236 236 122 122 212 212 129 131 130 130 23-5 104 104 121 120 62 63 108 108 67 66 I 66 66 From 285 to235 inches 513 515 590 591 306 306 531 531 325 326 325 325 Mean Results .
Gauge barometer .
Nitrogen .
Olefiant gas .
Air .
Carburetted hydrogen .
Fron 5 to 35 inches .
Tinme in seconds ... 514 306 531 325-25 From 285 to 235 inches .
Time ofoxygen= 1 0.8704 0'5182 0'9000 0'5512 For the experiments of the preceding table the form of the capillary H was so far altered , that a funnel-form was given to the apertures of the tube .
This was done by softening the tube in the blowpipe flame , within an inch of each extremity , and expanding the bore into a small ball of about one tenth of an inch in diameter ; the ball was afterwards cut across the middle , and left the tube of course with a cupshaped termination .
This change in the condition of the capillary seems to have no effect on the comparative rates of transpiration of the different gases .
It was found that the passage of air became a little slower by cupping the end of the tube by which the gas obtains ingress , in the proportion of 509 to 496 ; but cupping the point of egress occasioned no further change in the rate , the experiments being made within the pressures of 28'5 and 23'5 inches .
The transpiration rate of olefiant gas differs entirely from that of nitrogen and carbonic oxide , of which gases it possesses the theoretical specific gravity , and is a great deal more rapid ; the transpiration coefficient being so low as 0'5182 .
The specific gravity of the gas made use of was found to be 09840 , and it was absorbed by the perchloride of antimony to the extent of 96'5 per cent. The determination of the true coefficient for this gas is attended with unusual difficulty , from its constant and I believe unavoidable impurity , as prepared by the action of sulphuric acid upon alcohol .
607 XXIII.-Transpiration by Capillary H ( with cupped ends ) into a six-pint jar .
Baromn .
29'91 .
Temp. 54 ? .
Air .
Oxygen , Hydrogen .
Carburet .
hyd .
Carbonic oxide .
Carbonic acid .
Gauge barometer in inches .
I I. II .
II .
I. I. .
I. II .
I. 1I .
I. II .
28'*5 00000000000 26'5 212 212 234 233 105 105 133 133 206 204 175 176 24-5 212 212 235 236 105 106 134 132 204 205 176 174 23-5 108 107 121 119 53 53 66 67 106 106 90 90 From285S to23-5 inches 532 531 590 588 263 1 264 333 332 516 515 441 440 Mean Results .
Gauge barometer .
Air .
Hydrogen .
aCarburetted Carbonic Carbonic acid .
1a gebarometer .
Air .
Hydr .
hydrogen .
oxide .
From 28-5 to J Time in seconds ... 531'5 263-5 332*5 5155 440'5 23*5 inches Time of oxygen =1 0-9024 0-4473 0-5645 0-8752 0-7479 The results of the preceding table are remarkable as approaching more closely to the empirical numbers than any preceding results obtained by the same capillary .
The same observation applies to the results of the table which immediately follows .
In the one table , the number for carbonic oxide is 0'8732 , in the other 0'8752 ; the numbers for carbonic acid are 0'7479 and 0'7466 ; and for carburetted hydrogen 0*5645 .
TABLE XXIV.-Transpiration by Capillary H into a 6-pint jar .
Barom. 29'63 .
Temp. 55 ? .
Air .
Oxygen .
Carbonic oxide .
Carbonic acid .
Gauge barometer in inches..I IL .
I. IIH. .
I. I. II .
III .
I II .
1 I1 It II J / / I/ / / 28'5 000000000 26'5 i 212 212 236 235 214 214 213 175 176 24-5 214 214 237 239 203 205 205 178 177 23-5 110 110 21 120 100 101 101 90 91 From 28'5 to 23-5 inches 536 536 594 594 517 520 519 443 444 _~ ~~~~ ... .
I3 - .
Mean Results .
608 7ranspiration of different Gases by a Capillary Tube of Copper .
It appeared desirable to have experiments on the passage of gases through tubes of different materials , in order to ascertain how fiar the coefficients of transpiration observed are peculiar to glass .
After some trials , a capillary tube of copper was constructed , of a fine smooth bore , not wider than an ordinary thermometer tube , and indeed less in diameter than the preceding glass capillary H. This tube was formed by first drilling a cylindrical hole in the axis of a solid copper rod , 4 or 5 inches in length , and extending the latter afterwards by drawing it through a wire-plate .
An iron wire , or triplet , was placed within the copper tube and drawn through the wireplate at the same time , in order to keep the interior surface of the copper tube smooth and uniform .
It was necessary to pull out the iron-wire always after the copper was drawn through the plate , to prevent the former being fixed .
The iron-wire was then extended somewhat separately , and again introduced into the copper tube , and the operation of drawing out the latter repeated .
In this way the copper tube was extended to a length of 11 feet 8 inches .
It was found to be perfectly sound and airtight ; and allowed 1 cubic inch of air to pass into a vacuum in 22*12 seconds .
Of the iron wire , upon which the copper tube was last drnwn , 92-7 inches weighed 18'30 grains ; or 1 inch 0'1974 grain .
Taking the specific gravity of iron at 7'7 , this gives as the diameter of the copper tube 0*0114 inch , or gth of an inch .
When used as a transpiration tube it was coiled up into circles of about 10 inches in diameter , and the ends joined by soldering to two block-tin tubes provided with screws by which they could be attached to the aspirator-jar and drying tube .
The experiments were conducted precisely in the same way as with a glass tube , except that oil of vitriol was avoided and chloride of calcium only used in the drying tube .
The following series of results were obtained with the copper capillary .
TABLE XXV.-Transpiration by a Copper Capillary Tube into a one-pint jar .
Baroni .
29-97 .
Temp. 58 ? .
Air .
Oxygen .
Nitrous oxide .
Carbonic acid .
Air .
Gauge barometer in inches. .
_ I. II .
I. II .
I. II .
I. II .
( / ##a/ t/ t / , 28 00000000 20 234 233 260 261 198 198 199 199 234 12 291 292 324 324 244 244 245 245 292 8 206 205 228 228 170 170 171 170 204 4 333 335 372 373 275 277 76 276 336 2 X342 344 387 390 283 285 287 288 344 From 28 to 4 inches ... 1064 1065 1184 1186 887 889 891 890 1066 Mean Results .
Gauge barometer .
Air .
Nitrous oxide .
Carbonic acid mTime in seconds ... . .
233'5 198 199 From 28 to 20 inches .
, .
Time of oxygen 1 ... 0-963 7601 07639 .
Time in seconds ... ... .
291'5 244 2 45 From 20 to 12 inches .
From 20 t 12 nch Time of oxygen=l ... 08997 0-7531 0-7561 * Time in seconds ... ... 205-5 , 170 170-5 om t8 no Time of oxygen= ... 0'9000 0'7456 0'7478 { Time in seconds ... . .
334 276 276 From 8 to 4 inches ... ... oxygen=l 07410 7410 Time of oxygen0l..i96 0-7410 Time in seconds ... ... 343 284 287 From 4 to 2 inches ... ... of ygen= 08831 07310 7400 Timeofoxygen= 1 ... 0'8831 07310 0'7400 .
Time in seconds ... ... .
1064-5 888 890 From 28 to 4 inches ... ' Time in scnds ... ... From 28 to 4 iches " [ Time of oxygen=1 ... 0'8983 07493 07514 It will be observed that the numbers do not differ materially from those obtained with glass tubes , particularly with H. This is a capillary of great resistance , and therefore a deviation from uniformity of ratio may be looked for in the lower part of the scale .
A little irregularity in the rate of air , probably accidental , appears in the upper part of the scale ; but as with capillary H , the coefficient for air never varies far from 0*9 , at least between 28 and 4 inches .
Carbonic acid and nitrous oxide exhibit the usual parallelism of rate ; but in the upper part of the scale the excess of the coefficient above '75 is rather considerable .
Other experiments were made on the transpiration of carbonic acid into different aspirator jars , as the size of the jar and duration of the experiment appeared to have some influence on the ratios .
TABLE XXVI.--Transpiration by Copper Capillary into a one-pint jar .
Barom , 29'07 .
Temp. 56 , Oxygen .
Carburetted hydrogen .
Gauge barometer in inches..s ... ... .
_. .
: I. II .
I I. 11 28 0000 20 273 2 ; 74 1 151 12 338 3 336 i 186 186 8 237 2 236 129 130 4 383 385 210 211 2 391 400 014 213 From 8 to 4 inches .
11 131 676 678 610 Mean Results .
Gauge barometer .
Carburetted hydrogen .
__I XEnll : n ewi\r tvn wTime ' in seconds 1. .
15 From 28 to 20 inches T in seconds* .
151 Time of oxygen=1 .
05 0'5521 4Time in seconds ... ... 186 From 20 to 12 inches Time in second s.6 Time of oxygen-=1 ... 0'5519 vTime in seconds. . ?
... 129'5 From 12 to 8 inches ime in seconds .
12 lTimre of oxygean =1 .
0 . ?
05475 From 8 to 4 inches { Time in seconds , ... . .
210'5 Time of oxygen= 1 .
0*5482 rom 4 to 2 inch Tirne in seconds ... .
213*5 Fromi 4 to 2 'Inches rr .
/ ?
, / .
no Time of oxygen =1. .
0Q5398 Tinme in seconds ... ... 677 From 28 to 4 inches ime in seconds ... 77 Timne of oxygen= l ... 05500 The coefficient obtained for carburetted hydrogen in the preceding experiments never varies much from 0'55 , which is the mean between 28 and 4 inches .
The result is similar to that given by glass capillary H for the same gas .
TABLE XXVII.-Transpiration by Copper Capillary into a two-pint jar .
Temp. 56 ?
05 .
Barom. 29'49 .
Gau~ge barometer in Air .
Oxygen .
itrogen .
Carbonic oxide .
Air .
Hydrogen .
, , ]'/ / ?i : I : ?
/ ? ?
f ?
?t. .
i. Gauge baromete ? ? ?
?rur~ i , r in ------ , , ' , , , , inche 's .
11.1IL ~ I.--."J-.~-~____ 28 0 o0 0o o , 000 oi 20 421 419 467 466 405 407 405 404 ; 419 206 205 12 515i 514 572 74 501 500 496 498 516 255 254 8t 360 363 402 404 351 351 348 .
348 361 19 179 From 2 to 8 inches ... 196 1296 1441 1444 157 9 1250 1296 640 638 Mean Results .
Gauge barometer .
Air .
Nitrogen .
Carbonic oxide .
Hydrogen .
Fm 28 to 20 ich Time in seconds ... ... 420 406 404-5 205*5 From 28 to 20 inches .
'.^ ... .
; ~{ Time of oxygen= I ... .
09003 0'8703 0'8670 .
4405.F.o. .
to l che Time in seconds 514.5 500.5 497 j 254-5 ime of oxygen= 1 ... 08979 0*8734 0*8673 I 0*4441 From 12 to 8 iches Time in seconds ... ..361 .
5 351 348 179 rom to n Time of oxygen = 1 ... 078970 088709 08635 i 0-4441 F rom28 to8inchs { Time in seconds ... ... 1296 1257-5 1 49*5 1639 From 8 to 8 'inches Time of oxygen= 1 ... 0'8984 0'8717 0-8662 | 0'4429 The great resemblance which these results bear to those of the last glass capillary is most surprising .
The rates of air , nitrogen , carbonic oxide and hydrogen , may be considered as identical with these two capillaries , although they differ in substance , and also in the time of passage , which is slower in the copper capillary than in H , in the proportion of 22 ' 12 to 15'64 .
Carbonic oxide , it will be observed , is still sensibly more rapid in its passage than nitrogen .
The experiments in the preceding and all other tables are put down in the order in which they were made .
Tile observation with air is occasionally repeated , to find whether the rate of the capillary remains constant .
TABLE XXVIII.--Transpiration by Copper Capillary into a three-pint jar .
Barorn .
2973 .
Temp. 57 ? .
Guge barometer in Air .
Oxygenl Nitrogen .
Hydrogen .
Carbonic acid .
inches ... .
112 215 inches .
I II .
I. .
I I. I , .
Id IIr .
I fj / / / / .
/ it Jtf It !
I 28 000000000I0 20 595 595 663 660 575 577 287 87 500 502 12 736 735 818 819 713 713 358 357 616 613 8 516 517 T 575 577 500 501 251 252 426 428 From 28 to 8 inches ... 1847 1847 2056 2056 1788 1791 896 896 1542 1543 Mean Results .
Gauge barometer .
Air .
Nitrgeogen .
ydroen .
Carbonic acid .
Fv a t &gt; o 20 honk Time in seconds ... ... 595 i 576 287 501 From 28to20 nhes Time of oxygen= l. 0 ' 8994 08707 0-4339 0-7573 f Time in seconds ... ... .
73 55 713 357e5 614t5 From 20 to l2 inches Ti o xygen .
~Fro " tn2tlnc ~he Time of oygen1 ... . ?
S 0'8986 0'8711 0'4369 0'7507 Fr om 1to 8finhe{ Time in seconds ... . .
-5 505 2 427 rom 1 2to 8 Timeof oxygen=l ... 0894366 07413 From 28 to 8 inhes Timen seconds .
37 3S Time of oxygen= 1 ... 0'8984 0'8704 0'4358 0-7502 The preceding results with air and nitrogen might be confounded with those obtained with capillary H. The rate of hydrogen is sensibly faster , while that of carbonic acid is decidedly slower ; both gases diverging sensibly from their empirical rates 0'375 and 0'75 , on the side of effusion , but the former very slightly .
TABLE XXIX.-Transpiration by Copper Capillary into a six-pint jar .
Barom. 3014 .
Temp. 58 ? .
Gauge barometer in __ 5 ' _ydrogen inches .
I. II II I , IL .
IIL ~ IL , _E fw i\ , _ ... ... ... .7__ 28 00000 024 ... ... ... ... . .
614 269 271 20 1139 1139 1268 654 290 290 12 1413 1413 1570 1573 698 699 8 994 993 1109 1110 494 495 From 28 to 8 inches ... 3546 3545 3947 3951 1751 1755 I--1 1 111 111 _111 II , h 1 1 , ,llllllllllll , Hil._.l IIIII1 'i __ II , :r i_i iii iiiii Mean Results .
Gauge barometer .
Air .
Hydrogen .
Fr om 28 to 20 i nhe { Time in seconds ... 1139 560 Time of oxygen =1 0'8982 0-4416 From 20 to i nches Time in seconds ... ... 1413 698'5 Time of oxygen =1 0'8991 0'4445 From 12 t8 inches Time in seconds ... ... 993'5 494-5 Time of oxygen =1 0'8955 0'4457 m 28 to 8 in s Time in seconds ... ... 35455 1753 Fom 8 to 8 inch Time of oxygen --1 0'8978 0-4439 613 In these experiments with a large aspirator jar and long times , the rate of air continues very nearly 0-9 , and that of hydrogen approaches 0'44 .
The rates at the upper part of the scale are to be particularly attended to , as most uniform , and as representing pretty nearly transpiration into a vacuum .
TABLE XXX.-Transpiration by Copper Capillary into a six-pint jar .
Baroin .
30'23 .
Temp. 58 ?
'5 .
Air .
Oxygen .
Carbonic acid .
Gauge barometer in inches .
I. II .
I. IT .
I. II .
59 ? ?
59 ? ?
59 ?
'25 59 ? ?
59 ?
5 59 ?
-075 28 0 " 0tt 0 " 0 " 0"f 0 " 24 550 550 611 612 469 468 20 590 589 655 655 498 499 12 ... ... ... ... ... ... ... ... 1570 1570 1183 ' 1183 8 ... ... ... ... ... ... ... . .
1103 1104 824 825 From 28 to 8 inches ... ... ... ... ... ... ... ... .
3939 3941 2974 2975 7~~~~~~~~ 393 Mean Results .
Gauge barometer .
Air .
Carbonic acid .
From 28 to 24 inc , f Time in seconds ... ... 550 468'5 From 28 to 24 inches ... Time of oxygen= 0*8994 0 7661 Time of oxygen = 1.0'8994 0 7661 of Tlime in seconds ... . .
589*5 498*5 From 24 to 20 inches ... Time of oxygen= 1 ... 09000 07610 .
fTime in seconds ... ... ... ... ... 1183 From 20 to 12 inches ... 4 Time in seconds ... ... ... ... ... 8 From 20 to 12 inces ... Time of oxygen= 1 ... ... ... ... 0'7535 Time in seconds ... ... ... ... ... 824.5 From 12 to 8 inches ... ... ' ... .7472 Time of oxygen 1 ... ... ... ... 0'7472 From 28 to 8 inches f [ Time in seconds ... ... ... ... 29745 Time of oxygen= ... ... ... ... 0'7549 The preceding table contains an experiment on carbonic acid transpired into a large jar .
The coefficient of that gas still considerably exceeds 0'75 in the upper part of the scale .
It appealrs then that the rates of transpiration are similar through glass and copper tubes , for air , oxygen , nitrogen , hydrogen and carburetted hydrogen ; whilst the passage of carbonic acid is subject to a retardation in the upper part of the scale of the copper capillary , by which its rate increases so much as from 0175 to 0'7661 , in the last series of experiments .
The rate of carbonic acid indeed exhibits a want of steadiness with this capillary , which is not observed in the other gases enumerated .
Thus we find it in the upper part of the scale 0-7639 by Table XXV .
; 0-7573 by Table XXVIII .
; and 0-7661 by the last table .
I may add , that in a preliminary experiment which was made with this gas , and also in an experiment made subsequently to those recorded , and after the tube had been some weeks out of use , so high a coefficient was given for carbonic acid as 0'78 .
It is impossible to say whether this irregularity is properly referable to the material of the tube , or is peculiar to this individual capillary , as it is the only instrument of the same netal which was used .
The copper tube has no advantage over the glass capillary for experiments on transpiration , while it is liable to the objection that it cannot be used at all with certain gases which have a chemical action on copper , and would tarnish the surface of the tube .
The experiments made with it , however , have their value in demonstrating that the rates of transpiration of different gases are essential properties of these gases , and not regulated by the material of the transpiring tube .
Indeed there is no more reason to suppose that the coefficient of transpiration of a gas would vary with the substance of the tube , than that the specific gravity of the same gas would be found different according as it was observed in a glass or copper vessel .
6 .
Transpiration of different Gases by a Glass Capillary Tube E. This was another long capillary glass tube , resembling H , but somewhat shorter .
The extreme length of capillary E was 20 feet ; it allowed 1 cubic inch of air to pass into a vacuum in 12'03 seconds .
One inch of this tube at either end was found to contain 1 grain and 0-94 grain of mercury ; the smallest of which admeasurements gives 0,0187 inch as the diameter of the tube ; that is , about 3 th of an inch .
The following table contains a series of experiments on various gases , which were made on one occasion , and in the order in which they are given .
TABLE XXXI.-Transpiration by Capillary E into a six-pint jar .
Barom. 30'340-30-288 .
Air .
Oxygen , Carburetted Hydrogen .
Carbonic oxide Nitren .
Nitrous oxide .
Carboni acid .
Air .
Gauge barometer hydrogen .
in inches ... .
I. II .
I. I IL .
I II .
I. II .
I. II .
I. , I I. I II , .
I. I. I. 72 0720 72 ? ?
72 i25 720-75 720'75 720 75 7207575 720 75 720-75 72'75 720 75 72 ?
75 720'75 7275 775 720-75 7275 28*5 , .
0 " O " 0 " 0 " 00 0 " o 0 " '0 " 0 " 0 " 0I 0o 0o " 0 " Qt 0 : WI 0 26*5 153 153 -172170 93 93 75 74 150 149 150 149 3 129 129 130 130 154 154 24-5 156 156 X 173 175 97 97 75 76.150 151 151 153 133 133 132 132 158 158 23-5 84 84 92 92 52 50 41 40 80 80 81 80 70 7 70 70 84 84 From 28'5 to 23-5 in .
393393 437 437 242 240 3191 190 380 380 382 382 3332 332 33 332 396 396 43 7 4_x 614 Mean Results .
Gauge barometer .
Air .
Carbetted Hydrogen .
Carbnic Nitrogen .
Nitros Carbonic Air .
b hydrrogen .
r oxide .
oxide .
acid .
Frortm285 f Time in seconds 393 241 190s5 380 382 1332 '332 396 to 23'5 in .
Time of ox .
=1 0-8993 0*5515 0-4359 0-8696 0-8741 0-7597 0'7597 0'9062 This table exhibits the transpiration of the gases by E , in the upper part of the scale or into a vacuum nearly , the condition in which the subject is studied with most advantage .
The coefficients exhibit a close correspondence with those found by the two preceding capillaries .
The number for air is 0-8993 at the beginning , and 0*9062 at the end of the experiments ; a variation of rate to which this capillary appeared more liable than the others .
The number for nitrogen , 0'8741 , again slightly exceeds that for carbonic oxide , 0'8696 .
Nitrous oxide and carbonic acid have the same number 0'7597 , which is in excess compared with 0'75 ; while the number for hydrogen , 0'4359 , is slightly deficient compared with 0'4375 .
Carburetted hydrogen has the number 0'5515 , and is wonderfully constant with all these long capillaries .
As the time of transpiration in these experiments appeared rather short for exact results , a larger aspirator-jar was employed , and the experiment repeated with carbonic acid ; air and oxygen being added to give standards of comparison .
TABLE XXXII.-Transpiration by Capillary E into a nine-pint jar .
Barom. 30*136 .
Temp. 74 ?
05 .
Air .
Oxygen .
Carbonic acid .
Gauge barometer in inches .
-I .
II .
I. II .
I. II .
II It !
' t.X !
t Repetition of the last experiments .
Barom. 29'616-29'666 .
Temp. 72 ? .
Air .
Oxygen .
Carbonic acid .
Gauge barometer in inches .
I. II ... II .
II .
1 .
II / / / / 1/ / 28'5 000000 26,5 237 236 262 262 196 197 24-5 239 239 266 266 199 202 23-5 129 129 141.141 107 106 From 285 to 23'5 inches 605 604 669 669 502 505 Mean Results .
Gauge barometer .
Air .
Carbonic acid .
From 28'5 to 235 inches { Time in seconds ... ... 604-5 503.5 rTime of oxygen =1 ... 0-9036 0-7526 The three results for carbonic acid by this capillary are therefore 0'7597 , 0'7550 and 0*7526 ; of which the mean is 0*7558 .
The coefficient of capillary E for carbonic acid is therefore not so widely different from 0'75 as it at first appeared .
With these large aspirator jars the rise of temperature within the jar during the experiment becomes very small .
In the six-pint jar it did not amount to one-quarter of a degree FAHR. , and in the nine-pint jar it was altogether insensible .
I have now detailed the results of the transpiration of gases by all the capillaries used except one , -a tube with which a few preliminary experiments in the inquiry were made , not sufficiently precise to merit being recorded .
All of these tubes have given the same coefficient of transpiration to each gas , or coefficients closely approximating , although the tubes themselves have varied considerably in their respective dimensions ; namely , in diameter , from 4-th to Tylth of an inch ; in length from 2 inches to 22 feet ; and from 12 to 70 seconds in the time of transmission of one cubic inch of air into a vacuum .
It can be said that no selection of the tubes was made ; and their dimensions are in a great measure accidental .
From the agreement in results obtained with tubes so different in dimensions , I consider that a glass tube of any diameter whatever will be found suitable for observations on transpiration , provided a certain length is given to it .
If the tube is extremely short , a mere ring , then we know that gases will be transmitted by it into a vacuum according to the law of effusion ; oxygen , hydrogen and carbonic acid in times expressed by 1 , 0*25 and 1P176 respectively .
With the slightest elongation of the tube these ratios are disturbed ; the number for hydrogen soon increasing to 035 or 0'40 , and that for carbonic acid falling to 1 , or 0'80 , referring both to the time of oxygen always taken as unity .
The change with the increase of length is very great at first , but soon falls off , and with a certain length of tube seems to cease altogether .
The coefficient of hydrogen is then found to have risen to some number approaching closely to 0*4375 or 0'44 , and the coefficient of carbonic acid to approach closely to 0'75 .
The coefficient of nitrogen has fallen at the same time from 0'9373 ( its coefficient of effusion , that of oxygen being 1 ) to nearly 0'875 .
With this length of the tube , the influence of effusion upon the transpiration rate of the gas has ceased to be sensible .
The action of the tube attains a certain uniformity , at the same time , in another respect .
When the tube is deficient in length , the coefficient of the gas varies greatly with the extent of the exhaustion of the vessel into which the gas is flowing ; the rate of transpiration inclining most to the rate of effusion at a high degree of exhaustion of the aspirator jar .
But the amount of this variation progressively diminishes as the tube becomes longer , till at last the coefficient of transpiration remains nearly if not perfectly constant , whether the aspirator jar be entirely vacuous , or contains already gas of the tension of half an atmosphere , three-fourths or even seven-eighths of an atmosphere .
The tube is now of the length most favourable for experiments of transpiration .
Farther addition to the tube appears to have no effect in altering the coefficient of transpiration of air , provided the aspirator jar into which the air passes is vacuous or nearly so ; or the coefficient of transpiration into a vacuum appears to remain constant for all greater lengths of the tube .
But the extent of the barometric range in the aspirator jar to which the vacuum coefficient is found to apply is gradually limited .
Instead of extending over seven-eighths of an atmosphere it may be contracted to an eighth of an atmosphere or less , by greatly lengthening the tube .
The coefficients of transpiration which I have endeavoured to ascertain are properly therefore the relative times of passage of the gases into a vacuum , at the mean atmospheric temperature , or near that temperature .
But even with the most favourable length of the tube , when the aspiration is feeble , and does not exceed 2 or 3 inches of mercury , the rate of transpiration is modified from an interference , which is connected with the excessive resistance of the tube to the passage of the gas .
I have not been able to undertake an examination of the nature and extent of this interference , but suppose it to depend upon friction , while the rate of transpiration seems again to depend upon a constitutional difference in the gases themselves .
The theory of the transpirability of gases , which at present appears to me most probable , is that it is a kind of elasticity depending upon the absolute quantity of heat , latent as well as sensible , which different gases contain under the same volume ; and therefore that it will be connected more immediately with the specific heat than any other property of the gases .
The only other gases besides those already experimented upon , which could be retained over water , and exposed to the metallic parts of the apparatus without injury , are olefiant gas , nitric oxide and suiphuretted hydrogen ; and these were submitted to transpiration with several of the tubes already used .
I subjoin the results , although less complete than is desirable , under the head of each gas .
Mean Results .
Chayrburgetetned Olefiant te Gauge barometer .
Air .
Carburetted Olefiant 90 4+10OO .
hydrogen .
gas .
f Time in seconds ... ... 5215 318* 300 316*5 From 28-5 to 23-5 inches Time of air= 1 ... ... ... ... ... ... 06118 0-5763 0-6080 Time of oxygen= .
, ... ... 0-5506 0-5186 0-5472 619 The results for olefiant gas can be considered only as approximative from the circumstance that the gas was always contaminated by a small quantity of carbonic oxide , varying from 3 to 5 per cent. , and a sensible trace of the dense gas or vapour , to which allusion has already been made .
Both impurities would tend to raise the number for olefiant gas ; ten per cent. of carbonic oxide purposely added to that gas , raising its number from 0*5186 to 0*5472 .
The calculated mean rate of the last mixture is 0'5513 , taking the rate of carbonic oxide at 0-8700 , and that of olefiant gas at 0*5186 .
If the true coefficient for olefiant gas should be a whole number , it may be expected to be 0-5 , or exactly one-half of the rate of oxygen .
8 .
Transpiration of Nitric Oxide .
TABLE XXXV.-Transpiration by Capillary E into a one-pint jar .
Temp. 52 ? .
Air .
Nitric oxide .
Oxygen .
Air .
Gauge barometer in inches .
I. I. II .
I. I. , / , I , / / ' , '/ 28 00000 20 135 135 131 149 135 12 167 162 162 186 169 8 118 115 115 133 118 4 189 183 184 206 : 188 2 192 184 188 219 195 From 28 to 4 inches ... 609 595 592 674 610 Barom. 30'45 .
Mean Results .
XXXVI.-Transpiration by Capillary E into a two-pint jar .
Barom. 30'08 .
Temp. 60 ? .
Gauge barometer in Gauge barometeri* } Nitrogen .
Nitric oxlde .
Oxygen .
inches .
inches .
I. II .
, III .
I. ' II .
I. II .
I.X II .
I. IL - .
ii/ / ~~~~ O/ A / / / / / / / H/ 28 00 0000 20 231 232 ... ... 237 235 266 267 12 286 287 .
286 ; 288 287 327 328 8 200 200 : 202 203 206 227 227 4 320 317 320 313 312 363 ; 364 2 314 320 316 311 315 361 361 From 28 to 2 inches ... .
1351 1356 ... . .
1352 1355 1544 1547 Mean Results .
Gauge barometer .
Nitrogen .
Nitric oxide .
From 28tol2inches .
Time in seconds ... ... 517*83 523'5 Fromii28to l2InchesT ieof oxygenTime of oxygen= ... The experiments with this gas indicate a number above 0*60 for its coefficient of transpiration ; 0'625 is 5-8ths of oxygen , while the result of Table XXXVIII .
, which is the more valuable of the two , is 0'6161 .
But this gas is likely to vary sensibly in its coefficient with different capillaries , like carbonic acid ; and both its physical and chemical properties oppose difficulties to obtaining a correct result .
The rate of air following this gas is made very sensibly slower in the last table ; indeed the rate of the capillary appears to be permanently altered , possibly from the deposition of sulphur fitom the gas .
[ I shall add , in an Appendix to this paper , a series of observations on the transpiration of gaseous mixtures , which appear to warrant the following conclusions : It appears from Tables XLII .
and XLIII .
, that mixtures of oxygen and nitrogen in all proportions maintain a rate which is sensibly the arithmetical mean rate of the two gases .
From Tables XLIV .
and XLV .
, that the rates of mixtures of oxygen and carbonic oxide are also uniform , but that mixtures of carbonic oxide and hydrogen deviate greatly from the mean , inclining always to that of the heavier gas .
In Tables XLVI .
and XLVII .
, that the mixtures of oxygen and carbonic acid maintain the mean rate ; and that mixtures of carburetted hydrogen and hydrogen deviate greatly from the mean , always inclining to the rate of the slower gas .
Tables from XLVIII .
to LIII .
inclusive , exhibit the transpiration of hydrogen mixed with various other gases , particularly nitrogen , oxygen , carbonic acid , nitrous oxide , and nitric oxide ; and Tables LIV .
and LV .
contain the results of the transpiration of mixtures of carburetted hydrogen with oxygen and with hydrogen .
It appears that while all the other gases tried appeared to maintain their usual rates of transpiration in a state of mixture , those of carburetted hydrogen and hydrogen are greatly altered ; and when the proportion of the latter gas is not more than from 5 to 15 per cent. , its rate becomes as slow as the densest gas with which it is mixed ; the deviation from the mean in hydrogen mixtures being relatively greatest when one of the gases is present in a small proportion .
With an addition of so much as 25 per cent. of hydrogen , carburetted hydrogen , carbonic acid and nitrous oxide continue to be transpired in sensibly the same times as when pure and unmixed .
Hydrogen is then transpired of course as slowly as the other gas with which it is mixed , although the time of hydrogen alone is 0'44 , while that of carburetted hydrogen is 0'55 , and of carbonic acid and nitrous oxide 0175 .
Indeed small additions of hydrogen , such as 5 or 10 per cent. , made to carburetted hydrogen , appear to prolong the time of transpiration ; and what is very curious , raise the time of the mixture to the empirical number of pure carburetted hydrogen , namely , 0-5625 ( Table XLVII .
) .
A slight retardation of the same kind may also be perceived in the similar carbonic acid mixtures .
The transpiration-time of equal volumes of hydrogen and carbonic acid is 0'7339 , or very little less than that of pure carbonic acid .
622 Carbonic oxide and nitric oxide , with equal admixtures of hydrogen , correspond as closely in their times as when pure , as will be seen on comparing together the results of Tables XLV .
and LIII .
A similar apparent rectification of the time of nitric oxide will be observed , in the last of these tables , to be effected by the addition of 5 per cent. of hydrogen to that gas , as was remarked above of carburetted hydrogen ; the time of 100 NO2 being 08661 , that of 95 NO2+5His found 0'8788 ; while the empirical coefficient for nitric oxide is the same as that of nitrogen , namely , 0'8750 , or 0'8785 , adopting REGNAULT'S density of the latter gas .
It appears that the time of carburetted hydrogen is sensibly increased by the addition of oxygen , at least when the proportion of the latter gas amounts to or exceeds 25 per cent. of the mixture ( Table LIV .
) .
The times of transpiration of the hydrogen mixtures , which have been most minutely observed , namely , those of hydrogen with oxygen , with air , and with carbonic acid , are exhibited by the curves of Plate XXXV .
These curves start from a common point 44 , the time of pure hydrogen , and terminate respectively with the times of oxygen 100 , of air 90 , and of carbonic acid 75 .
It would be premature to enter at present upon any discussion of these results ; for the full elucidation of the transpiration of mixed gases , must await , I believe , the further extension of our knowledge of the laws of gaseous diffusion .
Nov. 1846 .
] 623 XLIV.-Transpiration by Capillary E into a one-pint jar .
Barom. 29'69 .
Temp. 54 ? .
Oxvgen .
Carbonic 97'5C0+250 .
95C0+50 90CO+100 .
75C0+250 .
Air .
Oxygen .
'Oxpgen oxide .
Gauge barometer in inches .
'_ I. I. , I ... I. I. .
I. tl , , 1 , ,t , , , , / I / / 28 00000000 20 157 137 139 138 140 142 143 158 12 195 169 169 170 172 176 175 196 8 137 118 120 120 120 124 123 138 4 219 192 : 191 192 194 198 197 221 2 221 190 192 193 202 202 199 221 From 28 to 4 inches ... 708 616 619 620 626 640 638 713 Mean Results .
Gauge barometer .
Carbonic 97.5CO+2-50 .
95C0+50.90CO+100 .
75C0+250 .
' Air .
Oxygen ... , .
K_ a _oxide .
From 28 to Time in seconds ... 616 619 620 626 640 38 4minhes &lt; Time ofoxygen -1 .
0-8701 0-8743 0*8757 0*8842 0 , 90400*9011 nce alc .
mean ... ... ... ... 0'8733 0-8765 0*8830 0*9026 Il ... ... ... ... ... .
Il Il. .
_ Ill ... .II ... ... ... ... ... ... ... ... .
TABLE XLV.-Ttranspiration by Capillary E into a Temp. 52 ? .
one-pint jar .
Barom. 30'18 .
Air .
Carbonic 66'7C0+33'30 .
50CO+500 .
66-70+33-3CO .
750+25CO .
Oxy 95CO+5H . !
90CO+10H.175CO+25H .
92'5CO+7'5H .
Air .
Hydro .
Gauge barometer oxdeg gel .
in inches . ---. .
I. I. eI .
II .
I. I. I. .
I. I.I. I. I. / / / // // / / / _ .
/ __ / / / // / /. .
28 0 V0 000 00 'O O0O0 20 137 132 140 143 146 148 152 133 132 128 132 137 78 12 169 164 172 177 181 183 189 164 163 159 163 170 84 8 119 116 122 125 128 130 133 115 116 113 115 121 59 4 192 185 195 200 206 205 215 184 185 181 185 193 95 2 193 188 201 205 209 204 219 190 191 181 191 '201 100:From 28 to 617 597 .
629 645 661 666 689 596 596 581 595 621 316,4 inches ... Mean Results .
Gauge barometer .
Air .
Carbonic 66'7CO+3330.s50CO+500 .
66'70+33'3CO .
750+25CO .
95CO+5H .
90CO+10H.75CO+25H .
92'5CO+7'5H .
Air .
Hydrogen .
Time in sec ... .
617 597 629 645 661 666 596 596 581 595 621 316 From Timeofox.= 1 0-8905 0-8664 0-9129 0'9361 0-9593 0-9666 0-8650 0-8650 0-8432 0-8635 0-9013 0-4586 28hto 4 Calc.mean , A* ... ... 0 ... ... ... 0.9109 0-9332 0-9554 0-9666 0-8046 08256 0-7644 0'8358 inches .
Calc.mean , B ... ... ... ... ... ... ... ... .
0-84501 0-8237 0-7848 0-8344

