Researches on Irritability of Plants

former position. If the temperature be raised again, there is once more a growing erection, and when the death- point is reached there is a sudden spasmodic contraction. But if the specimen once passes through the tempera- ture at which the spasm takes place, then there should be an abolition of all further response, proving the sudden contraction at 60° C. to be the spasm of death. Thus, after obtaining the sudden inversion of the curve at 60° C. in the last experiment, the plant was kept at that temperature for I5 minutes. Cold water was now substituted in the bath, and the record was taken once more of the effect of rise

Fic. 59.—Abolition of response to warming or cooling in specimen which had passed the death-point. and fall of temperature. A record is reproduced (fig. 59) which exhibits the result. In the lower curve is shown the record of effect of rise of temperature from 45° to 65° C., and in the upper the effect of cooling from 60° to 45° C. It is seen that while in the last experiment the plant exhibited a spasmodic contraction at 60° C., there is no trace of such an effect in the present case. The very slight movement observable in the two curves is the physical effect of heating and cooling, quite negligible compared with the physiological erectile movement due to warm- ing and the subsequent spasmodic contractile movement heralding the initiation of death-change.

In order to discover how constant is the death-point, I repeated the experiment with numerous other specimens. We have seen that the pulvinated organs present in the leaves of Desmodium gyrans and the bean plant (Vicia Faba) exhibit responsive movement under excitation. In fig. 60 is depicted the death record taken under standard conditions with the leaf of Desmodium. The record was commenced at 35° C.; it is seen that thermo-mechanical inversion took place at 61° C.

Fic. 60.—Record showing death-point of Desmodium ane (og (C- The next figure (fig. 61) shows the record with the leaf of bean plant. Here the responsive movements are very large. The inversion is seen to take place at 60° C. The occurrence of death-spasm may also be shown by means of ordinary plants. If we take a hollow tubular organ, such as the hollow leaf-stalk of gourd or hollow flower-stalk of any other plant, cut it in the form of a spiral and subject it to the rising temperature of the bath, there is noticed at first an expansive movement of uncurling of spiral. On reaching the death-point, however, the former movement is suddenly reversed to one of curling.

Flowers like French marigold exhibit death-spasm by sudden movement of opening or closure. found that there is induced an electric-spasm at the onset of death. When the temperature is rising, a given point of the plant-tissue exhibits increasing galvanometric posi- tivity, till at the critical temperature there is a sudden electric inversion into galvanometric negativity. The Fic. 61.—Thermo-mechanical inversion indicating death-point at 60° C. in leaf of bean plant.

electric-curve of death is of the same type as the thermo- mechanical curve. With specimens of Musa and Amaranth the death-point was found to be 59°5° C.1 As the death-spasm is a form of physiological response we should expect the curve of death to undergo modi- fication under physiological variation. One such modifica- tion would lie in the translocation of the point of inversion, or the displacement of the death-point. Thus age has some influence, the death-point of very young specimens being lower than that of mature ones. I shall demonstrate

! Bose: Comparative Electro-Physiology, p. 202. Longmans, London, the influences of other agencies, such as fatigue or poisonous drugs, in the displacement of the death-point. I have already given a record which showed the death- point of the leaf of bean plant to be 60° C. under normal conditions. Employing a similar specimen, fatigue was induced in it by means of tetanising electric-shocks ; the death record was then taken in the usual manner. It

Fic. 62.—Lowering of death- Fic. 63.—Effect of poison point under fatigue ; death- in lowering the death-point. spasm took place at 37° C. will be seen (fig. 62) that in this particular case, on account of fatigue, the death-point was lowered from the normal 60° C. to 37° C., that is to say, by as much as 23° C.)5 Iie lowering of the death-point, I find, is determined by the extent of fatigue. In order to discover the effect of poisonous solutions on the death-point, I subjected a specimen of the bean leaf to dilute copper-sulphate solution and took its thermo- mechanical record (fig. 63). The effect of the poisonous agent is clearly demonstrated by an appropriate lowering of the death-point, in this case from the normal 60° C. to 42° C., ot by 16° G,

The electric response of galvanometric negativity is characteristic of the living condition of the vegetable tissue. Dead plants do not exhibit this characteristic electric-response. When a plant is subjected for a time to a temperature of 60° C. its electric response disappears, such abolition being indicative of the death of the plant. A leaf of Mimosa subjected to abrupt variation of temperature—either sudden cooling or sudden warming— exhibits excitatory reaction. But if the temperature be gradually raised, there is a progressive erectile movement of the leaf ; gradual cooling induces a depression of the leaf.

When the leaf of Mimosa is continuously raised in temperature, then at a critical point the erectile expansive movement is suddenly converted into one of spasmodic contraction. This inversion takes place under standard conditions at or about 60° C. After this the response of the plant is permanently abolished. Various other plants, sensitive and ordinary, exhibit this characteristic death-spasm at or about 60° C. In taking an electric record it is found that an electric- spasm also takes place at the critical temperature, which is very near 60° C.

The death-point of the plant is lowered under physio- logical depression. Thus under fatigue induced by tetanising electric-shocks, the death-point was lowered from the normal 60° C. to 37° C. Poisonous reagents also lower the death-point. In a particular case poisonous solution of copper sulphate lowered the death-point by 18° C. Difficulties of accurate determination of Latent Peviod—Advantages of Resonant Recorder—Simultaneous tracings of tuning-fork exciter and Resonant Recorder— Automatic stimulation at a definite moment—lIdentical value of latent period in successive determina- tions—Accurate measurement of time-interval shorter than *oo5 second—Latent period little affected by inertia of recorder—Tabular statement of value of different specimens of Mimosa—Effect of season on latent period.

WHEN the motile pulvinus of Mimosa is subjected to an exciting shock, a short time elapses between the incidence of this shock and the initiation of the responsive movement. This short interval is known as the Latent Period. In a responding muscle, similarly, contraction does not occur instantaneously on the application of stimulus. The latent period in this case is determined from the record of the muscle-twitch. When after the application of stimulus the muscle has not yet begun to contract, the record appears as a straight line. Then on the commencement of con- traction, the recording-lever is jerked up and the curve likewise bends upwards. The length of the straight portion of the record, between a mark that represents the incidence of the shock and the flexure at the initiation of response, gives us the duration of the latent period. We have, however, to determine the time-value of this length. This is done by means of a sinuous curve drawn below the record by a tuning-fork, vibrating 100 or 200 times in a second. Experimenting in this manner, the latent period for frog’s muscle has been determined at about ‘or second.

As the muscle-record and the time-record are separate, certain error is likely to be introduced in inferring the time-value of any point on the muscle-curve (fig. 64). This error becomes relatively serious when the total time to be measured is very small. There is, again, the difficulty of exactly determining the point of flexure which represents the beginning of mechanical response. More troublesome still is the error due to the inertia of the recording-lever. On account of this and the mechanical inertia of the respond- ing muscle itself, the latent period thus obtained appears somewhat in excess of the true value.

In the apparatus which I employed, these difficulties have been reduced to a minimum. In _ the first place, the curve of response phytogram is at the same Fic. 64.—Latent period of hyoglos- time a chronogram. The sus muscle: aa, moment of error which might arise stimulation ; ab, latent period. from an inference based on Pata SPOR Dre SECOne a neighbouring time-record z is thus eliminated. I will later explain also the means that make it possible to determine the point of flexure, representing the beginning of the responsive movement, with relative accuracy. And lastly, the error due to the inertia of the recording part of the apparatus is reduced to a minimum by making the writing-lever excessively light. In the muscle recorders the weight of the recording- lever is about 3°5 grams. The lever which I employ weighs only ‘o4 gram. The recording part of my apparatus is thus nearly a hundred times lighter than that used for muscle records.

The accuracy of the time-record when made by the response recorder itself may be gauged from records giving simultaneous tracings of the exciting standard tuning-fork (roo D.V.) and the resonant vibration in the recorder induced by it. This latter had been previously tuned to give exactly 100 double vibrations in a second. A light aluminium stylus attached to the tuning-fork traced a sinuous line on a falling plate of smoked glass. The top of the vibrating recorder was so adjusted as to make suc- cessive dots during its vibration, simultaneously with the tuning-fork tracings. It will be seen from the record (fig. 65) that, corresponding to the crest of each tuning- fork wave and slightly to its right, we have a dot. The record given represents a period of fourteen one-hundredths of a second, there being fourteen crests made by the tuning-

Fic. 65.—Simultaneous record of vibrating-recorder and too D.V. tuning-fork exciter. fork time-marker, and exactly coincident with these are the fourteen dots made by the vibrating recorder. The interval between any two dots, therefore, is an accurate measurement of one-hundredth part of a second. If the plate be moving at a uniform rate, the interval between these dots will be uniform. But the accuracy of the time- measurements in the curve is independent of the rate of movement of the plate, for we calculate not by the distance but by the number of the dots. In the present figure the record was made on a plate which had been released and during its fall was acquiring increasing speed. The tuning- fork waves are thus gradually broadening out, and in exact correspondence to this the intervals between the dots are lengthening. When the phonograph motor which lets down the plate is just released, there is a short interval during which both that and the dependent plate are

acquiring increasing velocity. After this the velocity becomes uniform. If this uniformity should be_ required throughout the record, the tracing of response may be taken during this later period only. The mode of procedure, therefore, is first to make the recording-writer vibrate at its own definite frequency of, say, I00 times per second. The recording-plate is then released and later, when its motion has become uniform, we pass through the pulvinus an electrical stimulus of an instantaneous break-shock. There should be a mark made on the recording-plate corresponding exactly to the moment of stimulation. The horizontal record, consisting of a series of dotted points representing one-hundredth of a second, is suddenly deflected upwards on the initiation of the responsive fall of the leaf. The number of dots intervening between the mark of stimulation and this point gives us the value of the latent period for the specimen.

Stimulation cannot be effected by hand at any exact predetermined point on the record. This must be done automatically by the moving plate itself. We cannot again give an instantaneous break-shock without previously com- pleting the primary of the Ruhmkorff coil, which causes a disturbing make-shock. In order to avoid this the secondary electrodes, during make, should be short-circuited by means of a thick con- ducting-wire ; the secondary shock is thus practically diverted from the plant through the path of least resistance, which is the conducting wire. All these requirements are provided for in practice by the special mechanical devices of the apparatus.

The essential parts of the automatic arrangement by which a break-shock is given, at a predetermined point on the recording-plate, are shown in fig. 66. handle k. The winding disc is attached to the revolving axis of a phonograph motor. The disc is wound in a right- handed direction, which at the same time winds the spring Fic. 66.—Apparatus for determination of latent period of Mimosa: m, spring motor; w, winding disc; c, projecting catch; xk, release- handle, pressure of which also completes primary circuit of induction- coil; k!, short-circuit key. The automatic break consists of contact- rod adjusted by micrometer-screw A.

of the phonograph motor. The circumference of the disc is the same as the length of the recording-plate. One complete turn pulls the recording-plate up to its highest position. A projecting catch below the disc is caught by a pin attached to the spring-handle kK, when a complete turn has been made: the recording-plate is thus held arrested at its highest position. When desired, a pressure on the handle xk releases the disc, the axis of the motor begins to unwind, and the plate is allowed to fall. The motor is fully wound at the beginning, and the partial unwinding during one revolution is exactly compensated before the next observation by the winding necessary to pull up the plate. Owing to the constancy of this winding, the rate of fall in successive experiments is kept the same. The pressure of the handle k, which releases the plate, also causes ‘ make ’ of the current in the primary coil. This circuit of the primary coil is completed in addition through a contact-breaking device.

This consists of a long strip of ebonite, fixed along one edge of the recording-plate carrier. On the lower end of the ebonite a conducting-strip of platinum is sunk in and provided with a binding-screw. In front of this slides a rod with contact-point tipped with platinum. This can be adjusted up or down by means of a fine micro- meter-screw, A. When the recording-plate is released, carrying with it the conducting-strip, the primary circuit is broken as soon as the line of junction between platinum and ebonite is reached. This sudden interruption of the primary current gives rise in the secondary coil to an instantaneous break-shock, which passes through the plant. In order that shocks in successive experiments shall always be given at the same definite predetermined position in the fall of the plate, the following device is adopted: The recording-plate, as we have seen in a previous chapter, slides up and down a vertical support of triangular section. A movable peg fixed in the support holds it temporarily at a certain selected point chosen as that at which, during the descent of the plate, the shock is to be given auto- matically to the plant. For the purpose of adjustment a galvanometer is interposed in the primary circuit. So long as the point of contact-rod is in touch with the

conducting-strip, so long there will be a deflection in the galvanometer. By means of its screw-adjustment, the rod is gradually raised till the line of junction between platinum and ebonite is exactly reached; the deflection in the galvanometer will now cease suddenly. In this way the point of interruption or ‘ break ’ is determined with precision. By pulling the thread in connection with one arm of the recording-lever, we then trace a slightly curved line on the smoked plate. This indicates the exact position in succeed- ing records of the moment of application of stimulus. This mark of stimulation is shown in the printed records as a vertical line. After making this mark on the plate, the peg is removed. It is easy to see that in successive experi- ments stimulation will occur at that definite moment which corresponds to this marked line of stimulation.

K’ represents a key-device by which the make-shock is prevented from exciting the plant. One end of a lever carries a bent metal-rod of U-shape, which is partly immersed in cups of mercury by means of a spring. During the depressed position of this key, the secondary coil is short- circuited. When the handle k is slightly pressed, there is a ‘make’ of the primary current. But the make-shock is short-circuited as kK’ is still in the depressed position. Further pressure of the handle k lifts k’ up, removing the short-circuit of the secondary coil. When the break-shock is given by the contact-breaker of the falling plate, there is no short-circuit to divert the shock, which now passes through the plant and excites it.

By turning the disc p the recording-plate is lifted and held arrested in the up-position. The pressure of the handle xk releases the plate-carrier, which then begins to fall. At the same time, the primary circuit is completed and a make-shock is induced in the secondary. But this make-shock is diverted by the short-circuit key kK’, which is still in the depressed position. Further, pressure of the handle kK removes the short-circuit by lifting the ends of kK’.

All this takes place during the continuance of one pressure of the handle. During the descent of the plate, stimulation due to instantaneous break-shock takes place at the definite moment corresponding to the stimulation mark. Electric connections are appropriately made with the plant by means of threads moistened in dilute saline solution. One electrode of the secondary coil is thus connected with the stem of the specimen: the moistened thread in connection with the other electrode is lightly wound round the pulvinus. It is sometimes preferable, for reasons previously explained, to make this contact with

Fic. 67.—Two successive records, exhibiting identity of latent period. Recorder 100 D.V. per second. glycerin. The connections are so made that the current of the break-shock enters by the stem and leaves by the pulvinus, the latter being thus the kathode. We shall understand later the reason of this, in as much as the kathode is the point of excitation. I now describe the record of an experiment (fig. 67) carried out in summer for the purpose of determining the latent period in a specimen of Mimosa. Stimulus was applied at the point marked by the vertical line, and the upper of the two records was the first taken. The vibrating-recorder employed had been tuned to exactly 100 vibrations per second ; successive dots therefore represent intervals of

‘or second. It will be seen that the responsive movement begins to occur between the tenth and eleventh dots, and very near the latter. There are thus 109 spaces, each of the value of ‘or second, and the latent period is therefore ‘Iog second. In order to test to what extent successive experiments might give concordant results, I took a second record with the same specimen, which appears in fig. 67 as the lower of the two, having given the plant an interval of rest of 20 minutes after the taking of the first record. It will be seen that the second record is essentially a replica of the first, thus demonstrating that with proper precautions

Fic. 68.—Record of highly excitable specimen, taken with too D.V. recorder on a slowly moving plate. successive experiments on the value of the latent period will give results which are of extraordinary constancy. By making the travel of the recording-plate very rapid, the successive dots become more widely spaced and the minute time-intervals involved are made more conspicuous. But this has the disadvantage of rendering the flexure of the curve representing the responsive movement less abrupt, making the exact point of initiation of response somewhat more difficult to discriminate. Going, on the other hand, to the opposite extreme of making the travel of the recording- plate slow, the flexure of the curve becomes more abrupt, enabling us the better to detect the point of initiation of the responsive movement. The time-dots, however, are now closer together. This can be seen in another record (fig. 68) obtained with a vigorous specimen. Here the

number of spaces before the initiation of response is eight, the latent period being therefore ‘08 second. The closeness of the time-dots is not any great difficulty here, as with the help of a magnifying glass it is quite easy to make the necessary observation. For the determination of the latent period in plants this accuracy of an order higher than hundredths of a second is more than ample. But such a limit is easily exceeded. As an example of this, I give a record (fig. 69) made with a different recorder, whose frequency was an octave higher than the last—namely, 200 double vibrations

each second. The successive dots are therefore in this case xiz part of a second apart. It has been said already that by slowing the travel of the recording plate the abruptness of the flexure of the curve would be increased, the spaces between the dots being at the same time shortened. But we may obtain wider spacings without losing this sharpness of flexure, by making a magnified photo- graphic reproduction of the curve, as shown in the next figure (fig. 70), which is a reproduction of the first part of the record in fig. 69 enlarged about three times by photo- graphic means. In this way it is not difficult to measure, say, one-fifth of the distance between two successive dots, themselves representing an interval of ;1, part of a second. In other words, the calculation can be carried into thousandths of a second. In the present case there are