]\gt ; II .
On the Time-relations of the Excursions of the Copiltary , ?
nith a Description of the JTethod of Using it for the of of Short Duration .
GEORGE J. BURCH , B.A. , Oxon .
Communicated by ofessor BHOLOMEW PRTCE , F.R. Received September 3 , \mdash ; Read November 19 , 1891 .
[ PLATES 3-6 .
] IN a preliminary note " " On a Method of Determining the Value of Rapid Variations of a Difference of Potential by Means of the Capillary Electrometer I indicated briefly results of an investigation of the time-reIations of the excursions of the capillary eIectrometer .
The object of the present paper is to give further details concerning the method of measurement which I then put forward , and to describe the experiments which led to it .
The theory of the instrument has been treated by LIPPMANN , HERMANN , and others , mainly from a mathematical stand-point , and the phenomena have been ascribed to polarization , but no adequate attempt appears to have been made to establish by actual measurement the accuracy of the formula for the time-relations , nor to ascertain whether there were any other influences at besides those assumed in stating the problem .
POISEUILLE'S experiments on the velocity of the flow of liquids capillary tubes seem to have been overlooked , and no sufficienb investigation made of the effect of the form and dimensions of the tubes upon the of movement of the meniscus .
It seemed , therefore , that there was scope for a thorough examination of the problem by the experimental method , with a view , not only to the further elucidation of the theory , but also to the discovery of the best manner of using the instrument so as to render available those special properties which make it unlike any other form of eIectrometer or galvanometer .
In the course of the investigation , which has occupied more than three years , over a hundred electrometers were made , besides many additional capillaries used in preliminary experiments .
CXCII .
64 .
92 .
General Data with respect to thoe -Relahons of Excursion .
Dilution of the acid beyond a certain point was found to lessen the rapidity of an lrsion without altering the extent of it .
This indicated that the velocity of the movement was affected by electrical resistance .
A similar result was also produced by the introduction of 100,000 ohms into the external circuit .
( 2 .
) Shortening the capillary so as to reduce the length of the acid column made the instrument act more quickly , but in this case the mechanical friction , as well as the electrical resistance , was lessened .
The shape of the tube , where it tapers to form the capillary , was found to have a marked effect ; the motion of the meniscus being much more sluggish if this part was made too long .
In this case the increase of electrical resistance must be comparatively trifling , and the result is to be ascribed mainly to friction and eddy currents .
( 4 .
) The same may be said of the orifice , a clean-cut capillary being usually much quicker in its action than a splintered one .
These things may be taken as indicating the action of both mechanical friction and electrical resistance in determining the rate of movement of the meniscus .
The next step was to ascertain whether there was any latent period before the commencement of the rise or fall .
was done by photography .
A difference of potential was suddenly communicated to the instrument by striking open a short-circuiting key , the end of which projected across the slit upon which the of the mercury column was thrown , and so gave upon the sensitive plate , as it passed rapidly behind it , a record of the exact instant at which current was allowed to act upon the electrometer .
In no case could the smallest interval be detected between the opening of the key and the commencement of the excursion , even when the difference of potential employed was very small .
The of this was found to be true under ordina1y circumstances , that is to say , the meniscus ceases to move the moment the source of electromotiye force is withdrawn .
In other words , the electrometer is practically dead-beat .
It has been generally held be perfectly so , but this is not the case .
If the instrument is oue that has been specially designed to act with great rapidity , it will be found , on com municating to it a fairly strong charge from a condenser of not more than one-third of the capacity of the electrometer , that the meniscus will suddenly forward , and then slowly return , haps as much as one-tenth of the distance it has traversed .
But this overshooting is entirely prevented by the introduction of an external resistance of , say , 50,000 ohms .
Apparently the reason why this phenomenon has not been noticed is , that experimenters have contented themselves with observing the effect.of suddenly the circuit during an excursion ; but in so doing , they have mtroduced an inh'nite resistance .
In the case of the condenser there EXCURSIONS OF THE CAPILLARY ELECTROMETER .
is , as it were , a eservoir on each side of the electrometer , which acts as a spring , permitting a certain oscillation of the charge .
A far more severe test is to preserve the electrometer circuit unbroken , and to remove the source of electromotive force , by breaking the primary circuit of the rheochord .
This I have done , and the resulting photographs show the effect of overshooting to be , for the same velocity of the meniscus , nearly twice as great .
But it can still be oyercome by external resistance , and is noticeable when the difference of potential is two or three times as great as it ever is in physiological experiments , and even then I have not found it exceed 2 per cent. of the length of the complete excursion , with no external resistance .
It remains true , therefore , that with an ordinarily quick electrometer , under conditions which obtain in practical work , the meniscus to move the instant the source of electromotive force is withdrawn .
The velocity of the movement is ] greatest at the first , and there is no portion of the normal photographic curve , even at its commencement , concave to the asymptote , except in those cases in which oyershooting may occur .
Under such circumstances , slight signs of increase of the velocity are sometimes just discernible .
But this was only detected in three cases , when the currents used were as great as the electrometer would bear without injury .
These facts , implying the absence of acquired ' momentum seemed to indicate that in ordinary working the velocity of the meniscus at any moment must be some function of the accelerating force at that moment , in the sense that it is independent of any previous motion , and can contribute nothing\mdash ; or practically \mdash ; to the velocity with which it moves during the next interval .
In order to determine experimentally the form of this function , I arranged a rheotome in the derived circuit of a rheochord in such a way that any desired difference of potential ] be introduced into the electrometer circuit for an accurately measured period of from sec. to about sec. This rheotome , which was upon the principle of a drop-shutter , acted by first snatching open a short-circuiting key , and then after .
the required interval , breaking the circuit on both sides of the electrometer simul- taneously\mdash ; this arrangement being necessary , in order that the meniscus might stop suddenly without running back to zero .
A specially sensitive electrometer was used , having a long range , and with no tendency to ' creep It was less rapid in its action than those usually employed in the laboratory for physiological work , being selected on account so that the time-measurements might be more accurately made .
The insulation was so good that no appreciable return of the meniscus took place during 10 minutes after the circuit was broken .
The experiments were conducted as follows :Having found the total length of the excursion produced by a difference of potential derived from 150 millims. of the rheochord wire to be 126 divisions of the eye-piece micrometer , the rheotome was set so as to check the meniscus exactly in the middle of the excursion , namely , at 63 divisions from zero .
This required a closure sec. The rider of the rheochord was then ifted to 50 millims. , when ] the complete excursion was found to be 42 divisions , and that given the rheotome , 21 divisions .
On further reducing the length of rheoto millims. , the full excursion being 21 divisions , the meniscus shot up , when the heoton ) was used , of the eye-piece micrometer and is to , with only one-sixth part of the original difference of , the rniddlc point of the excursion reached in exactly the same time , namely 414 sec. In other words , the mean velocity of the meniscus during the first half of a normal excursion } to be proportional to the total length of it .
Further ents s the me was true of other fractions besides the half , as for , one-third , and three-fourths , but in these cases the measurements were less reliable , owing to purely instrumental difficulties .
From these data I concluded that the velocity with which the meniscus moved at any instant .
a normal excursion , must be proportional to its distance from the final position of at that time .
Fig. 1 represents the curve .
a normal excursion photographed upon a plate movlng with constant velocity horlzontally from to left .
The asymptote AB is taken as axis along the time is measured .
Let be a polnt on the curve vertical ordinate through P. , if denote the velocity of the meniscus in the capillary But , according to the experimental results , varies as EXCURSIONS OF THE CAPILLARY ELECTROMETER .
so that , whence Integrating or , It is obvious that any curve having this equation can easily be recognised in either of two ways .
The tabular logarithms of a series of ordinates corresponding to equal time-intervals , are in arithmetical progression .
This method was used in verifying the hypothesis as to the time-relations of the normal excursion .
The second is a graphic method depending on the fact that the subtangent NT ( fig. 1 ) of a logarith- mic curve , or intercept , upon the asymptote between the tangent and ordinate to any point upon the curve , is of constant length .
This property affords a means of finding the position of the asy mptote when only a portion of the excursion been included in the photograph .
Two points are chosen some distance apart , and the tangent and ordinate drawn to each .
The level is then found by trial , at which the horizontal distance between the one tangent and its ordinate is equal to that between the other and its ordinate , or , in other words , the level is found at which the subtangents are constant .
The possibility of doing this in the case of the normal curve suggested the method of analyzing other curves propounded in my preliminary note .
If , as my experiments have indicated , the velocity of the meniscus at any moment is due solely to the difference of potential between the terminals of the electrometer at that moment , and is not influenced in any way by its motion at any previous time , and adds nothing to its velocity at , ny subsequent time\mdash ; in other words , if the electrometer is perfectly dead-beat\mdash ; then the rate of movement at any given instant , during an irreguIar change of electromotive force , must be exactly the same as would be communicated to the meniscus by a permanent difference of potential equal to the electromotive force acting at that instant .
That is to say , the tangent to a given point on the curve produced by excursion must coincide in direction with that of { .he commencement of the normal curve given by the difference of potential still exishng between the of the electrometer .
If , therefore , the length of the constant subtangent to the normal curve is known , the length of the ordinate for that particular inclination of the tangent can be determined , and it will show how far the meniscus has still to move before reaching the position in the capillary corresponding to the difference of potential to which its velocity is due at the time in question .
If the mercury is at zero , the length of the ordinate thus determined will represent the actual electromotive force at that stant .
If it is not at zero , the distance it has aJ.ready travelled must be added ) or subtracted from the ordinate , according as its sign , and that of the ordi n , is positive or negative .
In this way an excursion may be yzed .
Points are taken on the curve at intervals of 00.50 , and the gent and ordinate drawn to each and produced , upwards if the curve is , and downwards if it is falling , until the horizontal disthem is to the ngent of the normal curve taken through the .
The ends of all ordinates are joined , and the curve so produced represents , on the scade as the ordinary readings of the electrometer , the difference of potential during each instant of the experiment .
Tt be useful to note that the ordinates of the original curve give the integral of the quantity of electricity that has passed at any instant .
' Investigation of for the of the Normal Excursion .
rtained , by measuring a of photographed excursions , that the fornl [ would hold for some instruments not for others , I proceeded to investigate the causes of these divergences .
There are two things which may modify the time-relations of the movement\mdash ; the one accidental , the other essential .
The first is that the instrument may not be of equal sensitiveness throughout the part used .
Obviously this difficulty may be got by .
a suitable electrometer .
The second source of error cannot be so eliminated .
It is that the internal resistance varies with the position of the meniscus the capillary .
( 1 .
) Error.\mdash ; This has a twofold effect .
It alters the electrical value of the xdings , and it has also a powerful influence upon the velocity of the movement .
The greater the range of the excursion for a given difference of potential , the slower is the action of the instrument\mdash ; not only relatively , but absolutely .
The relation between the sensitiveness of an electrometer , ond the time of half-charge , could not be determined directly by experiment , owing to the difficulty of making two pillaries sufficiently alike in all other respects ; but from a large number of observations it is evident that the time of half-charge increases much more rapidly the sensitiveness .
I found by experiment that the electrical capacities of electrometers , with capillaries of the same internal diameter , are in direct proportion to the of the excursions produced in them respectively by the same small difference of potential , and that the capacity is within wide limits independent of the difference of potential .
As , therefore , the same change of surface-tension has to force the mercury column through distances proportional to the sensitiveness , and a quantity of electricity has to flow which is also proportional to the same thing , seems probable that the time of half-charge may vary according to the square of the ) veness .
iments will be made with regard to this .
The important point , EXCURSIONS OF THE CAPILLARY ELECTROMETER .
established by a number of observations , is that the calibration error has a much effect upon the dity of the of the meniscus than upon the scale readings .
In most cases the sensitiveness of sn electrometer increases as the meniscus approaches the tip of the capillary ) and its of action becomes less .
But this calibration error is entirely under the control of the maker of the instrument .
( 2 .
) Changes of \mdash ; The principal seat of electrical is in the slender column of dilute acid between the meniscus and the end of the capillary .
But since the length of this varies continually during an excursion , .
follows that resistance must also vary , and with it , the velocity of the movement , must increase the meniscus the end of the , where the resistance is least .
This variation was investigated in a series of experiments .
A special electrometer was prepared , with a very long capillary , of exactly equal sensitiveness for a considerable distance from the tip \mdash ; twice as much as was required for the experiments .
It was extremely slow in its action , but was otherwise a very instrument , Fig. 2 .
To measure the time-relations of the the following simple and nsive rheotome was devised .
Upon a base-board , fig. 2 , clamped to the bench so as to project beyond it , are fixed two upright pieces , between which the bar swings until it comes in contact , as it assumes a vertical position , with the cross-bar , making connection in so doing between the terminals and , with which it is connected by a fine wire .
As it is necessary that the circuit should be broken on both sides of the electrometer , to prevent " " creeping a light wire , balanced by the counterpoiso is attached to , and moves with it .
This wire is bent so as to dip into two pools of mercury , in the block of paraffin , thus making a connection between the terminals and , about 01 sec. before the bar comes in contact with .
The weight whtch is a disc of lead weighing a couple of pounds , is attached , in the manner shown , to a string , passing through a loop on the end of the bar , over the rod on which it swings , to a cleat .
The weight being drawn forward and then let go , carries the bar up against the stop , and then swings under the , oscillating bout the loop at the end of , as a centre .
On its return , as it passes the vertical , the contact is broken , first at , which determines the period of closure , and immediately after , before the electrometer has time to creep , at , on the other side of the circuit .
It is then caught in the hand of the operator .
The closure may be maintained during any odd number of swings by pressing the thumb upon bhe bar , directly it has made contact , so as to hold it firmly against the stop , releasing it directly the weight has passed under the table for the last time , thus leaving it free to break contact on its return .
The duration of a single swing is calculated by taking the time of , complete oscillations with the aid of a stop-watch .
By suitably regulating the length of the string , the time of closure can be adjusted to any desired period from 4 sec. to 10 secs .
The accuracy of the instrument was by comparing the results obtained with different lengths of string .
The object of the experiments was three-fold , .
:\mdash ; ( A. ) To compare the time-relations of charge and discharge .
( B. ) To ascertain the rence , ifany , of the rate of movement when the meniscus was made to , i.e. , to advance towards the tip of the capillary , and when it was caused to fall , i.e. , to recede from the tip .
( C. ) To measure the effect of the change of internal resistance during an excursion .
As the instrument was one of constant sensitiveness , this could be done by comparing the ratios of successive ordinates corresponding to the equal time-intervals given by 1 , 3 , 5 , 7 , etc. , swings of the pendulum-rheotome , the mean of a number of observations being taken in each case .
The rheochord was set so as to give an excursion as near as possible of the full length of the scale of the eye-piece micrometer\mdash ; 50 divisions\mdash ; and the capillary was fixed so that a mark 20 divisions from the tip coincided with the top of the scale .
A low power was used , and the actual ran ge of the movement was much greater than is employed in physiological work .
One series of observations was taken with an added external resistarlce of 60,000 ohms .
For measuring the rate of discharge , the rheotome was made to close a short circuit between the terminals of the electrometer .
In the following experiments , distance of the meniscus from the tip of the capillary , in divisions of the eye-piece micrometer .
distance of the meniscus from its final position of rest after a closure of infinite duration .
The period of closure is expressed in swings of the pendulum rheotome .
The last column shows the ratio of each value of to the one preceding it .
' ' Up means an excursion towct.rds the tip of the capillary .
' ' Down signifies an excursion away from the tip of the capillary .
EXCURSIONS OF THE CAPILLARY 1 .
CHARGE .
No External Resistance .
Down .
Period of a single swing of the heotome sec.2 .
CHARGE No External Resistance .
Up .
Period of a single swing of the rheotome sec. 3 .
DlSOHARGE .
No External .
Down .
Period of a single swing of the rheotome sec. 4 .
DISCHARGE .
No External Resistance .
Up .
Period of a single swing of the rheotome sec. 5 .
CHARGE .
External Resistance ohms .
Up .
Period of a single swing of the rheotome sec. N.B.\mdash ; In the last three series of observations the period of the swing of the rheotome made longer , in order to effect a better distribution of the measured points .
On ] ) the results of.identical excursions , the CLlryes were found to coincide with those obtained ith a .
of 36 sec. , proving the accuracy of the method to be sufFicient for the purpose of these experiments .
The aboye experiments indicate\mdash ; That the time-relations of and discharge are practically alike , as is shown also by analysis of the photographed excursions .
That , in this particular instrument , the meniscus moved more quickly from the tip of the pillary than towards it .
No such difference could be detected in the case of electrometers of ordinary quickness of action , and the phenomenon appears to be peculiar to those of great sensitiyeness and slow action .
In some strun)ents it was reversed , and the mercury would only recede when assisted by tapping the , though it would advance freely with a small difference of potential .
That the change of the length of the acid column during an excursion a measurable effect upon the time-relations of the movement , which , however , is much less than it wouuld ) if the normal velocity were ditioned simply by the resistance of the electrometer : It was found by experiments with more than twenty ilJstruments that the effect of added external resistance in increasing the time of " " charge \ldquo ; was such , that they might be assumed to have an internal resistance of the form , where distance of the meniscus at any moment from the tip of the capillary , and is a constant of the particular instrument employed .
In practice , is many times greater than the maximum values of , so that the variation of the speed-ratio from this cause is less than 3 per cent. over the entire width of the negatiyes referred to in this paper , and consequently not more than per cent. in the physiological curves to which the method of analysis was applied .
To sum The two causes which modify the time-relations of an excursion , act in opposite ways .
EXCURSIONS OF THE CAPILLARY CTRO M The decrease of internal electrical resisbsnce as the meniscus } ) ches the ) the , tends to increase its velocity , while the of sensitiveness ) it move more slowly .
the latter has a much more powerful effect than the former , and may practically neutralize it , or even overpower it , without a calibl'ation error great enough to make an appreciable diffelence between the scale readi n 's and the true value of an excursion .
The time-relations of the movement are conditioned partly by electrical resistance , but mainly by some other cause\mdash ; probably mechanical friction .
The Production Anatysis of Photographs of the Excursions of the Capiltary Electrometer .
The rapid movement of the sensitive plate required to out the details of the electric phenomena of muscle and nerve , necessitated an alteration in the form of the apparatus .
The ingenious arrangement deyised by the Rev. F. G. SMITH for producing a rectilinear motion of uniform velocity was inapplicable , owing to lack of space in the dark room .
I therefore made an apparatus , the details of which be described at length elsewhere , in which the sensitive plate was caused to describe an arc of a circle .
The dark-slide containing it was tached to a kind of balanced pendulum , which carried it , at a uniform velocity , past the slit on which the magnitied image of the column of mercury was thrown .
The requisite velocity was given to the pendulum by a weight which , as in ATWOOD'S machine , was caught by a stop j before the plate reached the slit .
The return of the pendulum after the exposure was prevented by a catch , and the key producing the excursion was actuated at the right moment by an arrangement of electromagnets .
Time was recorded upon the plates by a magnetic vibrator placed in front of the slit , and driven by a tuning-fork in the usual way .
The exact instant of excitation was recorded by the signal-key before referred to , which was also ) , in front of the slit .
With this arrangement the normal curve is most easily expressed in polar coordinates .
Time being recorded upon a circular arc , becomes Instead of the rectilinear asymptote , there is an asymptotic circle of radius The expression for the radius vector is the equation connecting and being This change necessitated a reconsideration of the method of analysis .
The curve is a variety of arithmi i , but the cept between the tangent and the radius vector on the asymptot , ic circle\mdash ; the line corresponding to that which I had employed in the case of the rectilinear motion , is not constant , and not of a form convenient to use .
There is , however , another property of polar coordinates which is at once available ; namely , that the expression for the polar sub-normal is In this case so that this property enabled me to analyze the curves with remarkable ease , and I propose to abandon rectilinear motion for the plates in Fig. 4 .
The problem of the analysis of a photographed resolves itself therefore into the determination of the length of the subnormals of a sufficient number of po nts upon the curve , together with the measurement of the distances of these points from the circle corres o to the position of the meniscus for zero otential .
These measurements are most easily made upon the instrument re resented in fi .
3 .
presen The ative is fixed to acarrier , pivoted at , on the base-board , in which a * See ' Differential Calc.ulns , ' 7th edition , p. , from which I gob the clne while trying to elaborate a roundabout method of reproducing the polar curwe in rectangular coordinates , so that I go might analyze it .
EXCURSIONS OF THE hole is cut so that the photograph can be examined by transmitted .
The radius of the carrier is exactly equal to that of the pendulum on which the plate was exposed , and a fine wire or piece of ehair is stretched from to close underneath the glass without touching it , which serves as a radial line of reference .
An index fixed to the carrier , passing over a scale upon , enables small angular displacements of the carrier to be accurately measured , thus determining the time intervals , and the ordinates are found by laying a graduated rule upon the negative just over the radial line , using as a reference circle the edge of the ) which is always well defined .
The of the subnormal is found by means of the flat rod , , which carries at one end the glass plate , , shown 011 a larger scale in fig. 4 .
On this plate is ruled a fine line , , continuous with the edge of , but broken for a distance of about 2 millims. at about the middle .
Through the gap thus left passes the tangent line , , exactly at right angles to .
When this line , at the part where it is intersected by , is placed as a tangent to the curve which is to be analyzed , at the point where it is cut by the radial in , the length cut off by the rod , , upon the gradu ted rule , is permanently fixed at right angles to , is the subnormal to the curve at that point .
The plate carrier is then shifted through an angle corresponding to a known interval of time , as determined from the time record upon the photograph , and the subnormal measured again for the part of the curve thus brought over the radial line , .
This process is repeated at sufficiently close intervals throughout the curve , the corresponding ordinates being measured , taking the edge of the photograph as a reference circle .
For the application of this method it is first necessary to analyze a normal curve produced under the exact conditions , as to resistance , of the experimental photographs .
With a suitable instrument , .
the subnormal is practically a constant multiple of , the distance of the meniscus from its point of rest , and the value of this constant multiplier has to be found .
The following example of the analysis of the normal curve of the electrometer used in the physiological experiments described by Professor BURDON SANDERSON in a paper on the " " Photographic Determination of the Time-relations of the Changes which take place in Muscle during the period of so-called Latent Stimulation ( ' ' vol. 48 , p. 14 ) , will serve to show how this is done .
In this particular case , as the curve was obtained for the purpose of analyzing the physiological photographs which we had taken , the circuit was led through the non-polarizable electrodes and the muscle exactly as it had been arranged for those experiments .
An equivalent metallic resistance was then substituted for the preparation , and several other photographs of the normal excursion were secured .
But the resulting curves were found to give identical results and , therefore , the details of this one only are inserted here .
First , the ordinates corresponding to time-intervals of 001 sec. were measured with an ivory rule , graduabed very finely in millimetres , the tenth of a millimetre ' esf , imated with the aid of a lens .
As check , three different reference lines were taken on successive days , and the results so obtained were found to be identical .
The next of the operation consisted in the determination of the asymptotic circle .
In so short a period as that comprised between the across the slit of an eight-inch plate , the mercury has 1lot time to reach its final position , which must therefore be found by calculation .
Assuming for the time-relations of the excursion the simplest formula , given in my preliminary note , we may write for the distances from any reference circle of three points corresponding to equal intervals of time , where the distance of the reference circle from the asymptotic circle , and is some function of Whence ; ; three points as far apart as possible , I found the position of the asymptotic circle and reduced the observed readings of the ordinates to the values of measured from it , which were as follows:\mdash ; EXCURSIONS THE CAPILLARY ANALYSIS of Normal Cnrve No. 289 , fig. 5 , Plate 4 .
Upward exc , ursion equivalent to the difference of potential given by 83 centims. of the rheochord wire with : 1 cell .
Resistance in circuit , a gastrocnemius preparation with nonpolarizable electrodes .
Total length of the excursion millims. The column of calculated values is a series in geometrical progression , inserted between the first and the last terms .
It will be seen that the error is in all cases well within the limit of accuracy of the measurements .
Having thus determined the position of the asymptotiG circle , i.e. , the position of the meniscus when , on the assumption that the formula was exact , the next step was to try whether any traces could be discovered of the influence of change of resistance , or of the sensitiveness of the capillary , during the excursion .
In other words , I had to determine whether these errors were due to inaccuracies of measurement or of the formula assigned to the curve .
If the latter were the ca , se , then there must be some value of which would make increase or decrease with some approach to regularity from one end of the curve to the other , and we should have approximately , , so on .
On assuming trial values for , however , the discrepancies seemed only to increase , and , after a somewhat laborious investigation , I concluded that in this particular instrument the effect of diminished resistance was practically balanced by d ) slight increase of sensitiveness towards the tip of the capillary\mdash ; the one tending to increase , and the other to diminish , the velocity of the iscus , so that the equation represents very closely the time-relations of the excursion .
The process * These ten points correspond to intervals of 01 sec. , as given by ths time-tracing .
That is why the first one is some 3 mil-lims .
from the commsncement of the curve . .
G. J. BURCH ON TIM -RELATIONS OF THE above described is necessary with every new electrometer , in order to determine its constants .
When the equation of the curve has thus been determined once for all the is much more simple .
The position of the asymptotic cllcle found by direct snrement upon a screen placed in the of the sensitive plate , and a known value is taken through the resistance used in single normal excurslon nown the experiments to which method of analysis is to be applied .
Two or three ordinates with their corresponding subnormals are then measured , and the ratio of each ordinate to its subnormal is mined The mean of these ratios is taken as the value of the constant multiplier .
rement of the bnormal of Normal Curve No. 289 .
curve was spoilt for this purpose by the presence of small undnlations in two places caused by some jar to the apparatus ; these only interfered with the meant of at one point , where , however , it was easy to take the mean position ween them .
But in order accurately to place the fine ] upon the glass plate of the measuring rod as a tangent to the curve , it was necessary that this should be perfect for some little distance on each side of the point of contact .
The two most favourable positions were the beginning and the end of the curve , and these gave for the value of respectively Two other positions , not so well situated , gave and .
The remaining readings were less reliable on account of the undulations referred to , and a short piece where the definition of the raph was defective .
The mean of the four measurements is ; but is probably nearer the true reading .
With this instrument , therefore , a difference of potential due to 83 centims. of the rheochord wire gave an excursion of millims. on the sensitive plate .
With a in circuit equal to that of an ordinary physiological preparation , the subnormal to the curve at its commencement was centims. Wherlce 1 centim. on the subnormal centims. on the rheochord , and 1 centiIn .
on the rheochord ntims .
on the subnormal .
Comparison with other Excursions .
In order to show that the method may be relied on to give constant results , the following experiments are quoted .
The circuit was the same as for Curve No. 289 , EXCURSIONS OF THE CAPILLARY ELECTROMETER .
and the value of each excursion , as measured by its subnormal , may be compared with that obtained by calculation .
No. 290.\mdash ; Rider of the rheochord set at 67 centims. Muscle-current centims. , giving an effective difference of potential of 25 centims. Subnormal at the commencement of the excursion centims. Calculated 8.3 : 25 : : : No. 292.\mdash ; The difference of potential was simply that due to the muscle-current , which had fallen to 37 centims. The excnrsion was therefore in the opposite direction .
Subnormal at the commencement of the excursion centims. CalcnJated 83 : 37 : : These results agree fairly wel ] .
For greater accuracy , the actual measurement of the subnormal was made in all cases at a point from 2 millims. to 5 millims. from the commencement , and the difference of the ordinates of the measured point and of the zero ] , multiplied by the constant factor , added to the result .
Hence the number of places of decimals .
( See Part II .
, p. 101 .
) The muscle was then removed from the circuit , and the following experiments made .
No. 294.\mdash ; Upward excul.sion ( towards the tip of the capillary ) given by 80 centims. of the rheochord wire .
No external resistance .
Subnormal at the commencement of the curve centims. No. 295.\mdash ; Upward excursion given by 80 centims. of the rheochord wire .
External resistancs obms .
Subnormal at the commencement of the curve centims. No. 296 .
Downward excursion given by 80 centims. of the rheochord wire .
External resistanco ohms .
Subnormal at the commencement of the curve centims. These results show the identical character of the upward and downward excursions , and also that the introduction into the circuit of 1000 ohms makes no very marked difference ; they do , however , indicate the effect of taking away the resistance of the muscle and its electrodes , in that the value of the constant multiplier has risen from to about Artcfficial Spikes .
This name was given to the curves obtained by photographing the excursions produced by means of two currents in opposite directions , each lasting about sec. It was intended to imitate as nearly as possible the effect produced by the excitation of muscle , with differences of potential of known value and duration , so as to determine whether the capillary electrometer was capable of distinguishing between a current of definite strength suddenly thrown into it , and a more or less gradual rise of a difference of potential , extending over a period of equal duration .
That it can do so was shown beyond a doubt .
To produce these artificial spikes , the wires from the electrometer were joined up through the derived circuits of two rheochords , A and , each of which was provided with a separate battery .
These were conn ected with a rheotome furnished with four adjustable break-keys of the tip-over type .
The first of these , when struck , opened a short circuit on the primary of , allowing the current to act upon that rheochord .
The second broke the battery .
circuit of at the same instant the third in like manner opened a short circuit on the primary of the second rheochord , in which the current circulated in the opposite direction .
About sec. later the battery of was thrown out of action by the fourth key , thus the excursion produced by the second current .
It will be observed that the electrometer circuit remained unbroken throughout the experiment .
Unfortunately the short-circuit contacts were not good , so that a certain amount of current passed , rendering it impossible to ascertain the exact value of the differences of potential which acted upon the electrometer , in several cases\mdash ; and , nloreover , the Daniell 's cells used were not sufficiently steady during the operations , which took a long time , to enable all the results to be compared with those previously obtained .
The experiments were varied by reversing the direction of both currents , and sometimes the second .
No. 306 .
Fig. 6 , Plate 3.\mdash ; " " Artificial Spike produced by two short currents of opposite direction .
Up , 0055 sec. ; ) normal centims. Down , 0045 sec. ; subnormal centims. It will be observed that the effect of the second current was to bring the meniscus back almost exactly to its former level , and that both the rise and fall commenced and ended suddenly , indicating , as do also the measurements of the subnormal at close intervals throughout the curve , that it was produced by the introduction of differences of potential which remained while they lasted .
But the movement was in each case about one-tenth of the full excursion for that difference of potential , and the quantity of electricity received as " " charge viewing the electrometer as a condenser , is approximately proportionate to the product of the time into the difference of potential , or what comes to the same thing , the subnormal , .
:\mdash ; ' the upward movement .
for the downward movement .
But these " " charges\ldquo ; may be measured in another way .
The electrical capacity of electrometer is perfectly definite in amount , being conditioned by the shape and size of the ] ] , and perfectly independent of the difference of potential .
quantity received by an instrument from any current is therefore proportional to the product of its capacity ( which is constant throughout the part of the capillary used ) into the distance , through which the meniscus is moved by it .
The effect of two successive charges of equal quantity , but opposite in sign , must be to leave the mercury ultimately at zero , as was the case here .
Not only was there this close correspondence between the values obtained in two ways for the " " charge but Total Indicated Differences of Potential , calculated from the subnormals , were within one per cent. of the values found by the two rheochords employed , with that used for Curve No. 289 .
EXCURSIONS OF THE CAPILLARY ELECTROMETER .
The following curves are given as examples of the effects of ooting .
The actual difference of potential employed was not ascertained owing to the instrumental difficulties referred to , which were not noticed till the conclusion of the experiments .
No. 305 .
, Plate spike , given by an upward excnrsion lasbing sec. , followed by a downward excursion lasbing sec. The second phase is slightly rounded at the beginning , sbowing that the full velocity was not attained immediately .
The slight pause at the end of the fall , before the slow return to zero , is probably due to " " stickiness\ldquo ; of the tube .
( Compare also No. where there is evidence of " " stickiness\ldquo ; during the swiftest part of movement ) .
No. 807 .
Fig. 8 , Plate 3.\mdash ; Artificial spike , as before , but in the reverse direction .
In this case , the second phase begins sharply at its full velocity , but at the end of it the mercury has continued to move , not so rapidly as when the current was on , but still , more quickly than it would have done had the electrometer been simply discharging itself .
No. 309 .
Fig. 9 , Plate 3.\mdash ; Excursion produced by a single current lasting sec. Here the meniscus overshot the mark to the extent of millim. in an excursion , the full height of which would have measured 80 mlllims .
on the , and which had only risen 7 millims. when the current was checked .
The analysis shows that the second rheochord somehow became connected and furnished a current in the opposite direction of about one-fourth of the difference of potential of the first\mdash ; possibly one of the leads might have touched the rheotome .
It will be noticed that even this overshooting is under 1 per cent. of total normal excursion , and that the time which elapsed between the withdrawal of the electromotive force and the cessation of the movement , was only about sec. These three photographs illustrate the form of overshooting , which I consider to be due to the elasticity of the meniscus , for , if it were true overshooting of the column , then the retardation of the return movement should be visible in the case of the reversed spike ( No. 307 , fig. s ) ; but this was only seen with much stronger currents .
On the other hand , the sudden checking of an upward excursion , about three times the size of this , caused the meniscus to throw off a globule of mercury .
In the preceding experiments , there was no external resistance .
The effect of introducing a coil of 10,000 ohms is shown in the next example .
No. 299 .
Fig. 10 , Plate 3.\mdash ; Downward excursion produced by a single current of short duration .
External resistance ohms .
In spite of the difference of potential being greater than in any of the preceding experiments , there is practically no overshooting , and the curve of discharge is perfectly normal .
The subnormal to the curve was too great to be measure MR. G. J. BURCH ON LATIONS OF machine .
The slight notch about the middle of the descent , indicates a " " sticky\ldquo ; place in the capillary .
Compare fig. A companion photograph , in which the excursion was upward , did not develop sufficiently dense to print , but was similar in character , the upward overshooting being very slight .
These photographs show , that without external resist in the , the effect of overshooting , and especially of that form of it which is due to the elasticity of the meniscus , interferes seriously with the application of this method of analysis , unless the rences of potential concerned are small , but that these difficulties on the introduction of a moderate resistance , not sufficient to make the movements of the electrometer gish .
With suitable precautions , the variations of a difference of potential may be determined for each two-thousandth part of a second , and under favourable circumstances , with an error of not more than one per cent. PART II .
Application of the Method to the Study of the Etectrical Variations of Muscle .
In a paper by Professor BURDON , entitled " " Photographic Determination of the Time-relations of the Changes which take place in Muscle during the Period of so-called ' Latent Stimulation ' " " ( ' Roy .
Soc. Proc vol. 48 , p. 14 ) , it was stated that in the gastrocnemius of the Frog the electrical response to an instantaneous stimulus , as investigated with the aid of the capillary electrometer , is indicated by a sudden movement of the mercurial column of very short duration , and that the photographic expression of that movement shows that between the contacts two electrical changes of opposite sign , and not more than one two-hundredth of a second in duration , have immediately followed each other , or , more explicitly , that the spot excited became for about sec. first negative , then for a similar period positive , to the other contact .
This statement relates exclusively to the case in which the effect is led off to the electrometer by two electrodes of which one ( f ) is placed on the tendon , and the other ( m ) is on the belly of the muscle .
In this case , the photographic record of the electrical response to a single excitation of the nerve , resembles the record obtained , when in a circuit of the same resistance as the muscle , two currents of the same duration follow one another in opposite directions .
This resemblance between the two cases is , however , only general , for whereas in the artificial effect the difference of potential between the two terminals of the electrometer remains the same during each phase , changing sign at the moment that the first is followed by the second , the difference of potential between the two spots to * [ This notch has not be reproduced in the figure .
It can easily be seen in the negative even without a lens .
March 3 , 1892 .
G. J. B. ] EXCURSIONS OF THE CAPILLARY CTROMETER .
which the electrodes are applied gradually rises to a negative maximum the first phase , is reversed , and then , having become positive , gradually subsides .
My present purpose is to show in what way the method of analysis scribed can be applied to the photographic records of the electrical response ( which I shall hereafter speak of as the record or curve ) , so as to determine the exact time-relations of the changes above described .
With this view I propose to present to the Society certain records , and to state the results obtained by the analysis of each .
This result will be best expressed by a graphic denoting the actual diffel ences of potential between ( m ) and ( f ) at successive times during the period of electrical change .
It may be convenient at this point to describe in detail the method of analyzing such records as muscle curves , in which the electrical variations are of brief duration and not necessarily of constant intensity or of the same sign .
It has been shown in the first , part of this paper that it is not difficult to procure an electrometer of which the time-relations may be expressed by the formula and that , with a circuit of suitable resistance , the movement of the meniscus will commence the instant that a difference of potential is communicated to the and cease the instant it is withdrawn .
such au instrument under such conditions , as was stated in my preliminary note " " On a Method of Determining the Value of Rapid Variations of a Difference of Potential by means of the Capillary Electrometer\ldquo ; Proc , " " the velocity with which the meniscus is moving at any instant is that with which it would start if the zero-line were moved to the level then occupied by the meniscus , and the difference of potential existing at the time between the terminals of the electrometer were suddenly introduced and made permanent .
Thus , the total indicated difference of potential is made up of two parts , viz. , the difference represented by the distance through which the meniscus has already moved , and that indicated by the velocity with which it is still moving In the analysis of muscle curves , as will be seen , the latter is usually many times greater than the former , except at the beginning , middle , and end of the curve where the tangent is at right angles to the radius vector , and the subnormal is zero .
The method of analysis set forth in my preliminary note is , in consequence of the change in the apparatus already referred to , whereby the sensitive plate is made to describe an arc of the circle instead of travelling in a str ight line , superseded by the far simpler one given in the present paper .
The subnormal to the normal curve\mdash ; and , consequently , to any curve\mdash ; is a constant multiple of the distance of the meniscus from the point of rest corresponding to the difference of potential existing at the time between the terminals of the electrometer .
As has been already said , the value of this constant multiplier is influenced by the resistance of the circuit in each experiment , and the first step is to determine it for the particular resistance used .
This was done for the musclu curves of which the MR. G. J. BURCH ON THE LATIONS OF THF , analysis is appended , the normal excursion photographed for the purpose being No. 289 , and the value of the constant multiplier was The next step is to place the glass negative of the muscle-curve upon the measuring ( fig. 3 ) , carefully centreing it so as to ensure its occupying a position exactly esponding to that in which it was placed on the pendulum of the photographic apparatus .
If this is not done , the results will be falsified .
The rate of movement has then to be determined .
When , as in the cases given here , the angle subtended by the entire excursion is small , no appreciable error is introduced by employing a tangent scale instead of measuring the arc .
This being so , a very simple method is available , whereby the hundredths of a second , given by the time-record on the photograph , can be subdivided into thousandths of a second .
The index of the carrier ( fig. 3 ) , is a edge fixed radially , and the tangent scale , which is divided into millimetres , is placed at such a distance from the centre , that ten of its divisions shall be passed ovel by the index when the carrier is through a space corresponding to one-hundredth of a second , as indicated by the coincidence of two successive apices of the time-record on the photograph with the radial line The scale having been fixed in this position , the actual measurements are proceeded with .
The instant of excitation , marked by the upward movement of the arm , is taken as the zero of time .
This portion of the photograph is brought over the line , and the reading of the tangent scale is noted .
The distance of the electrometer record at that time from circular edge of the photograph is measured ; this gives the position of zero potential .
The next thing is to find the time at which the meniscus began to move .
For this purpose , the rod is laid along the line , and the line aa , on the glass plate , placed accurately as a tangent to the curve .
The carrier is then shifted until a point is reached when the curve commences to rise , and the time at which this takes place is noted on the scale .
Next , the carrier is further shifted in the same direction through a space corre sponding to sec. There are now two operations to be perlorrned:\mdash ; To measure the subnormal of the curve at this point .
The of doing this has been already described .
The result may be written Subnormal To measure how far the meniscus ] moved from its original zero position for This is done by taking the distance of the curve at this point from the edge of the photograph , which is used as a reference circle , and subtracting the result from the zero distance ah.eady ascertained .
The difference may be written Movement It is most convenient to express the total indicated of potential in terms EXCURSIONS OF THE CAPILLARY ELECTROMETER .
of the subnormal\mdash ; reducing it subsequently to decimals of a volt .
The total indicated difference of potential will therefore be Attention must be paid to the of both and .
In the examples given , an upward excursion is produced by the negative variation , so that the formula for the first phase will be At the apex the subnormal passes through zero , but the total indicated potential is not yet zero , being indicated truly by the distance of the meniscus from the zero line , i.e. , the actual height of the excursion .
This , expressed on the same scale as before , is During the second phase , or descending portion of the curve , the subnormal is positive , and we must write If , as sometimes happens , the meniscus should descend below the zero line , also becomes positive , and we have It is unnecessary to describe minutely the arrangement of the physiological preparations .
The photographic curves of which the analysis is given are some of those obtained in the experiments in which I assisted Professor BURDON SANDERSON , to whose kindness I am indebted for permission to make use of them in this , the first application of the method of analysis which this paper describes .
Non-polarizable odes were employed , one ( f ) placed on the tendon , and the other ( m ) on the belly of the muscle , unless otherwise stated .
The muscle was kept retched by a weight , not greater than could be lifted by a moderate contraction .
The exciting electrodes were applied to the nerve , the exact instant of the stimulus being marked by the upward movement of the arm of the signal-key , the shadow of which is recorded on the plate .
This key , which has been referred to , was placed in the primary circuit of the induction-coil , and by its , all possibility of delay in registering the instant of the induction shock was eliminated , save that due to the induction coil itself No. 287 .
Fig. 11 , , Plate 4.\mdash ; Gastrocnemius of R. esculenta .
Hixciting electrodes plaoed at the extreme end of the prepared nerve , as far as possible from the muscle .
The muscle curlent was balanced by the difference of potentiaI derived from 42 centims. of the wire of the rheochord .
The analysis of this curve is given in fig. 11 .
The excursion is not large , but the negative variation is sudden , and remains tolerably constant for some time .
The positive variation is not great , but the return to the original zero is slow .
The relative value of the ] current is shown by the dotted line , which represents the trne zero potential .
Immediately aftel ' this expcriment , without a lterin the electrometel cirouit , the normal excursion , No. 289 , wa\circ ; e phed for the determination of the constant multipJier used in this and the remain .
ing illustrations .
No. 270 .
ocnemlus .
electrodes as far as possible from the muscle .
12 b.\mdash ; The same muscle .
Exciting electrodes as close as possible to the muscle .
No. 2/ 1 .
Fig. 12 , same No. 272 .
same muscle .
xciting electrodes as as possible to the muscle as in No. 271 , but the loading-ofl electrode shifted from its previons position to a point 12 millims. nearer the tendon electrode ( f ) .
These three curves were analyzed , and the results plotted together in fig. 12 , Plate 5 .
It will be observed that the two former , though somewhat different in the first has , are fairly alike in the second , the delay of the electrical response , in consequence of the reater length of nerve through which the excitation was transmitted in the case of No. 270 , being well shown .
It is interesting to note that the analysis of No. 272 approaches more nearly in shape to that of an " " artificial spike The rise of the electromotive force is fairly rapid , and it remains nearly constant for a considerable part of each phase .
But neither the rise nor the fall is so sudden as in the case of Curve No. .306 , although , as the difference of potential is less and the resistance of the circuit greater , any error due to overshooting of either kind is quite out of the question .
No. 274 .
, a.\mdash ; Gastrocnemius .
Exciting placed on the , close to its insertion in muscle .
Leading-off electrode ( f ) ched to the tendon ; on the side of the muscle , nearly at its thickest part .
No. 276 .
Fig. 13 , .\mdash ; The same muscle .
Exciting eltodes as before .
Leading-off electrode ( m ) shifted into the nol'mal position on the belly of the muscle .
No. 277 .
Fig. 13 , same muscle .
Exciting electrodes as before .
electrode ( m ) placed so close to the tendon eIectrode as to give a scal'cely visible excursion of the meniscus upon excitation .
The analyses of these three curves were plotted together in fig. ] 3 , Plate 6 .
They all cross the zero line within a remarkably small space .
The first two differ slightly in the negative phase , but are tolerably similar in the positive , which is well developed .
In both there is a second rise of the positive variation , reaching a maximum about sec. after the excitation .
This rise is , however , too small to be discernible from the photograph , and is made manifest only by the method of analysis .
It does not come my province to discuss the physiological questions arising out of these results , which I desire to leave entirely in the hands of Professor BoRDON SANDERSON , to whom I am deeply indebted for permission to make use of these photographs , taken in the course of the research in which I had the honour of assisting him , in order to illustrate the method of analysis which I now bring forward .
My object has been to show that the capillary electrometer can be employed to obtain measurements which hitherto could only be effected by a process open to some objections .
The repeating rheotome , by Professor BERNSTEIN for the same purpose , EXCURSIONS OF THE CAPILLARY ELECTROMETER .
although it has many advantages , can only give the average time-relations of a number of e]ectrical variations due to a rapid of stimuli , and is inapplicable in any case in which changes may occur in the preparation from fatigue .
my experience of rheotomes leads me to believe that it is much more difficult than is sometimes supposed to obtain very short closures of constant duration .
The photographic records given by the capillary electrometer are free from both these objections .
They show the result of a siugJe stimulus or of a number in rapid succession , and circuit unbroken and therefore constant .
Hitherto no means by which these indications could be quantitatively interpreted , except in the case of slow changes , has been in the hands of experimenters .
I have endeavoured to supply this lack , and in doing so , to represent the capillary electrometer in its true light as a measuring instrument , not indeed equal to the galvanometer in delicacy , nor in durability , nor , perhaps , in accuracy ; but far excelling it and all other electrical measuring instruments in the one property which is peculiar to it , namely , rapidity of action .
Probably in this way the electrometer may be made to serve other purposes than those of physiological research ; and I foresee that this method of analysis may be applicable to other instruments than the electrometer .
All the photographs and many of the experiments described in this paper were made in the Physiological Laboratory , Oxford , and my thanks are due to Professor BURDON SANDERSON for his help and counsel , without which I could not have completed the research .
I have also to thank Mr. C. CRUMP , of the Record Office , for his valuable assistance in discussing the mathematical parts of the problem .
The complete method of ] the corrections for calibration error and change of resistance I had hoped to give with his aid , but it was found that some further experiments would be necessary to furnish the requisite data .
The treatment of this part of the subject therefore must be deferred .
In the instrument which was used for nearly all the photographs these errors were not sensible .
[ NOTE.\mdash ; The curves were analysed by measuring the atives , and the results plotted from left to right .
The photographs having been reproduced as prints , must be read in the reverse direction , from right to left .
March 3 , 1892 .
G. J. B. ] .
\mdash ; A. - ' Phil. ' { 4 ' - q ) .
1892 .
1 .
LL c '

