The Physiology of the Ascent of Sap

The motive power is thus assumed to be the root- pressure, supplemented by the backwardly transmitted negative pressure caused by transpiration from the leaves. Confining our attention to the latter, it would follow that the greater the partial vacuum produced by transpiration the greater would be the backwardly transmitted suctional force and the corresponding enhancement of the rate of ascent. As regards the channel of conduction of water, it is considered as certain that the xylem alone subserves this function ; the well-known ' ringing experiment ' is supposed to offer conclusive proof : ' In order to break the con- tinuity (of the cortex) two circular incisions are made round the stem right into the wood and the intervening ring of tissue removed. If this " ringing " be not done too extensively, and if due care be taken that the stem does not become dried up or rotten at the region of ringing, the leafy crown will remain fresh for a long time, and the transport of water will not be interrupted to any appreciable extent by the ringing. We may conclude therefore that the conduction of water is effected by the wood.' ^ This

experiment is by no means conclusive, since the injection of water into the xylem (see Chapter XII) by the active cortex below would carry the water through the short stretch of the woody tissue fiom which the cortex had been removed. Another argument adduced in support of this view is the supposed abolition of ascent of sap in a stem when its cut end has been exposed for a short time to the air : ' When a stem has been cut across, air is drawn into the opened tracheae and the tracheids, owing to the internal negative pressure, and hence the absorption of water is rendered more difficult. In herbaceous plants the lessened rate at which the water is then absorbed is sufficient to cause a pronounced fiaccidity even when the cut stem is immediately placed in water. ' ^

Thus, according to the generally accepted theory, the ascent is mainly due (i) to the root-pressure together with the internal negative pressure and backward suction due to transpiration ; it follows from that theory (2) that the greater the condition of drought caused by transpiration, the quicker should be the rate of ascent ; and (3) that the ascent should be stopped by previous exposure of the cut end of the stem, the vessels being choked with injected air.

The theory of cellular pulsation asserts, on the other hand, that (i) the ascent is due to the independent activity of living cells which extend throughout the length of the plant, hence neither root-pressure nor transpiration is essential to the process ; (2) the propulsion being due to cellular activity, which is enhanced under increased internal pressure, the rate of ascent should be diminished under condition of drought ; and that (3) it is not the dead vessels, but the living tissue which takes an active part in the conduction ; hence the previous exposure of the cut end of a stem to air should not cause a stoppage of the ascent. I proceed to describe experiments which prove (i) that the ascent may take place at a vigorous rate in the complete 1 Pfeffer, Plant Physiology, English translation, p. 231.

absence of root-pressure and transpiration, (2) that the rate of ascent is diminished under increasing drought, and (3) that the exposure of the cut end of the stem to air does not abolish the conduction of water in the stem. Ascent of Sap in the Complete Absence of Root- Pressure and of Transpiration I took a specimen of Chrysanthemum which had been subjected to incipient drought : its root and all but a single indicating leaf were removed. The stem and this single leaf were coated with vaseline for the complete elimination of transpiration. The specimen was duly mounted, and a fresh cut made at the lower end of the stem, to which water was applied by raising a beaker of water from below. It will be seen (Fig. 14, a) that the erectile response of the indicating leaf took place two dots, that is 30 seconds, after the application of water at the cut end ; the intervening length was 15 cm., and the velocity of ascent was thus 300 mm. a minute, or 18 metres per hour. In certain other instances the velocity was found to be as high as 70 metres per hour.

It is obvious that this high rate of conduction could not possibly be due to slow osmotic action. Moreover, in the experiment just described, there was no root-pressure to propel the water, nor any transpiration to suck it. Hence it follows that it is the cellular activity throughout the length of the stem which causes the propulsion of sap. Experiments on the effect of increasing drought in depressing the rate of ascent will be given in full detail in the next chapter : I here give a summary of some of the results. In the stem of Chrysanthemum, the average rate of ascent under moderate drought was found to be

230 mm. per minute. This was depressed under excessive drought to 18 mm. per minute, or to about one thirteenth. In Impatiens, the average rate in cut stems was 70 mm. ; under excessive drought this was depressed to 7-5 mm. per minute, or to about a tenth. For this experiment, a shoot of Chrysanthemum subjected to drought was taken and its cut end was exposed to air for more than half an hour. The xylem-vessels would then be filled with air under atmospheric pressure which would block the channels. I also removed all the leaves except the solitary indicator, and smeared the stem and the leaf with vaseline, thus producing a complete abolition of transpira- tion. There could now be no backwardly transmitted suctional force, nor was there any channel for conduction through the xylem, now choked with air. According to the current theory, there should be a complete abolition of the ascent of sap under the particular circumstances described above. According to the pulsatory theory, however, there should be no such abolition ; the pulsations of the semi- dried cells at the cut end would, it is true, be arrested ; but this arrest would not be permanent. For after the absorp- tion of water there would be a slow revival of activity ; the record would thus show a prolonged latent period followed by an ascending curve less erect than that of plants in which drought was not so pronounced.

The record given in fig. 14, d shows that an ascent of sap did take place along the stem in which the cut end had been previously exposed to air, and from which the transpiring leaves had been removed. The response-record, moreover, shows the characteristics which were expected. The latent period is prolonged to eight minutes, and the relatively slow rate of ascent is found in the gentle slope of the curve of the erectile movement. Thus by two independent tests we arrive at an identical conclusion that the ascent of sap takes place not by physical transference along the dead xylem, but along the living cells by means of their pulsating activity. Other experiments will be described in a subsequent chapter which will offer independent proof of the underlying physiological action in the transport of sap. •

The experiments described above prove that the xylem is not essential for the ascent of the sap. I have been able by an electric method to localise the tissue which by its pulsatory activity maintains the ascent of sap (see Chapters XIV, XV). This is the cortex which abuts on the fibro- vascular tissue. In dicotyledonous stems there is thus a cylindrical sheath, which subserves the purpose of rapid conduction of sap. The inactive xylem- vessels are situated very near the active cortex, within a fraction of a millimetre or so ; hence it is easy for the active cortex to force the sap laterally into the xylem during the phase of contraction. The xylem may, therefore, be regarded as a reservoir, water being pumped into or withdrawn from it according to the different circumstances.

It will also be shown in the chapters referred to above that the fundamental mechanism in the ascent of sap is the same in herbaceous plants and in tall trees. Additional means, however, become increasingly necessary to meet the excessive demand for water in trees during active trans- piration. In herbaceous plants the distance of the soil- water is not too great : but in tall trees it is necessary to have a near source of supply of water, a ' soil-extension,' as it were, in the shape of conduit-pipes filled with water. These conduit-pipes are the young xylem-vessels (alburnum) for mechanical transference of water during the emergency of active transpiration from the leaves. Physical forces alone, such as capillarity or the cohesive power of water-

columns, cannot raise water to any great heiglit. There is, however, no such limit in the case of propulsion of sap by the physiological action of living cells. When transpiration is feeble, the normal ascent along the cortex supplies every portion of the tree with water. The leaves become turgid, and the xylem filled with sap. During active transpiration, however, the physiological conduction is not sufficient to meet the demand, and water is withdrawn from the xylem-reservoir. Two factors are thus brought into operation : the physiological conduction by and along the active cortex, and physical transference along the xylem.

Drooping leaf and drooping stem become erected in consequence of restoration of normal turgor by the ascent of sap after irrigation. The automatic record of the erectile response gives an indication of the rate of ascent of sap. According to the generally accepted theory, the ascent of sap is due to root-pressure and suction exerted by tran- spiring leaves, the rate of ascent increasing with increased drought and transpiration. Conduction is supposed to take place exclusively through the X3dem ; if this were so, the ascent would be stopped by exposure of the cut end of the stem, since the vessels would then be choked by the injected air. In disproof of these views the following facts have been experimentally demonstrated.

The ascent of sap takes place with great rapidity in complete absence of root-pressure and transpiration from leaves. The ascent of sap persists in stems whose cut ends have been previously exposed to air. The above experiments prove (i) that the xylem is not essential for conduction, (2) that neither transpiration nor root-pressure is essential, and (3) that there are channels other than the xylem for the ascent of sap. Difference of velocity of ascent in cut and rooted specimens — Influence of the previous history of the plant — The Duplex Method — The effect of drought — The effect of physiological anisotropy induced by stimulus — Determination of velocity in the reverse direction — Summary.

In the last chapter two methods for obtaining records of the rate of ascent of sap were described. The results ob- tained by the application of these methods with the highest degree of accuracy will now be given, dealing with such questions as whether the velocity is the same in intact plants with roots and in cut stems ; the effect of increasing drought on velocity ; the effect of stimulus on the rate of ascent ; the effect of physiological anisotropy in inducing differences of velocity of ascent on the two sides of an organ ; and, finally, the velocity of movement of sap in a direction opposite to the normal.

In the determination of the velocity in different speci- mens, I was at first greatly puzzled by the widely divergent values obtained, which ranged from 0-3 mm. to about 700 mm. per minute. A long course of investigation enabled me, however, to detect the causes of this divergence : I found (i) that the velocity was not the same in different species of plants, the velocity in Chrysanthemum, for example, being higher than in Impatiens ; (2) that the velocity was higher in thick than in thin specimens ; (3) that it depended on the temperature, a rise of temperature up to an optimum enhancing the rate of ascent ; (4) that the velocity was higher in a cut stem than in a specimen with roots ; (5) that

the rate of ascent is modified by the previous history of the plant ; (6) that it is affected by the condition of drought to which it had been subjected ; (7) and that it is modified by the action of stimuhis, the after-effect of which may be persistent. With regard to the velocity of ascent in specimens with roots, the following table gives results which I obtained with potted specimens of Impatiens in a condition of moderate drought : Table V. — Velocity of Ascent of Sap in Specimens of Impatiens with Roots, 30° C.

The velocity in Impatiens in a ' moderate ' condition and at a temperature of 30° C. is thus found to be of the order of 10 mm. per minute. This velocity is lower than in cut specimens of Impatiens, which was found to be about 60 mm. per minute. The reason of this difference is found in the fact, already stated, that in specimens with roots, the fine root-hairs offer great resistance to the entrance of water. In normal specimens of Impatiens, i.e., those subjected to moderate drought, the velocity of the water-transport has been shown to be of the order of 10 mm. per minute. In a particular specimen, however, the result was found to deviate greatly from the normal. The latent period was very short, which meant a great enhancement of the rate of transport of water, which was independently exhibited

by the marked steepness of the curve of erectile response. In attempting to discover the cause of the anomaly, I found that the specimen had been warmed artificially to hasten the drooping of the stem, and this must have necessarily raised the temperature of the soil. The specimen, however, had been kept in the experimental room for several hours before the commencement of the experiment, by which time the temperature had returned to the normal. It thus appeared that the warming of the soil had stimulated the roots, the after-effect of which persisted even after return to the normal temperature.

In order to put this surmise to experimental test, I took three batches of similar plants which drooped to the same extent from drought. The first batch was kept as the control, the temperature of the soil of the second batch was raised, while that of the third was lowered. For producing vari- ation of temperature of the soil, two boxes were made with circular openings, through which the conical pots were let down, so as to close the opening, the pots being exposed to the air of the chamber. The temperature inside one of the boxes was raised by an electric heating coil, and that of the other lowered by fragments of ice placed at the bottom. The soil in the two sets was thus brought to about io° C. above and below the normal. It should be noticed that it was not the plant as a whole, but the roots embedded in the soil, that were subjected to the action of variation of temperature. After this the pots were kept in the experimental room for several hours till they attained the normal temperature, as indicated by a thermometer imbedded in the soil. Records were next taken after irrigation with water at the ordinary temperature, which revealed in a striking manner the dif- ference in their past history. The control batch gave the characteristic records which have been previously described, the average velocity of the transport of water being about 10 mm. per minute. The batch whose roots had been stimulated several hours before by warmth, now gave records which exhibited a very short latent period and a

velocity which was more than ten times the normal. The batch whose roots had previously been cooled exhibited no response for a considerable length of time, sometimes not even for hours. The velocity in such cases was about 0"3 mm. per minute, or one-thirtieth the normal. It is thus seen how profoundly the activity of the ascent of sap is modified by the previous history of the plant. We shall see later how the depression of irritability of the root affects other activities of the plant.

It is necessary here to explain certain difficulties which are encountered in the accurate determination of the velocity of ascent. These arise (i) from the loss of time due to the physiological inertia of the responding leaf or cut portion of the stem, (2) from the loss of time required for absorption of water by the root, (3) from the difficulty of measuring the intervening distance from the root, since its exact position with its numerous side-branches is very indefinite. These difficulties are, however, eliminated by the Duplex Method of record, where two indicating leaves, situated vertically one over the other, give successive responses to the arrival of water at the two points (see fig. 11). The response of the lower leaf indicates the moment of the arrival of the ascending sap at that leaf ; the delay in the response of the second gives the time-interval for the ascent of sap from one leaf to the other. The physiological inertia of the two leaves being about the same, this source of error is eliminated by taking the difference of the two latent periods. The question of the distance of the root does not arise at all in this method of determination.

We have, however, to bear in mind the characteristics of the leaf-arrangement on the stem. The leaves in Chrysan- themum are arranged in a spiral, so that the fifth leaf is situated vertically above the first. A particular ascending fibro-vascular bundle, moreover, gives off lateral branches to the first, the fifth, and the ninth leaf counted in order from below ; the vertically situated leaves are thus connected with each other. Another important fact is, as stated previously, that it is the cortex abutting on the fibro- vascular strand which is mainly concerned in the ascent of

Fig. 15. Diagram for Determination of Rate of Flow of Sap in Up and in Reversed Direction In the left diagram there is a vertical slit in the stem shown by dotted line ; a piece of mica is inserted in this slit. Water applied at the slanting cut surface x travels upward from A to B to c, after which it crosses over to d, and follows the reverse course d, e, f. Diagram to the right shows reverse direction of flow by irrigation of the leaves. A is the normal direction of the ascent, and d the reversed direction of flow.

sap. It thus follows that a particular cortical strand of tissue goes straight up, supplying the sap to the first, the fifth, and the ninth leaf. If the root of an intact plant or the cut end of the stem be irrigated, we can observe the successive erection of the vertical row of the leaves. A, B, C, or F, E, D, on the two opposite sides, and thus determine the velocity of ascent between A and B or between B and C and so on (fig. 15). I find that the velocity is approximately constant in the middle portion of the stem, which is neither too old nor too young. This will be seen in the results of the follow- ing experiment on the determination of the velocity in a specimen of Chrysanthemum which had been subjected to drought. The distance between a and b was 97 mm. and the time-interval between the successive responses was forty-five seconds. The velocity was therefore 130 mm. per minute. The distance between the leaves b and c was 62 mm., the time-interval thirty seconds, and the velocity 124 mm. per minute. The above results show that this differential method enables us to determine the velocity with great accuracy.

The following experiments demonstrate that the velocity is decreased under increasing drought. The results given in the accompanying tables may be regarded as typical of the effect of slight and of excessive drought on velocity. The specimens employed were cut stems of Chrysanthemum and of Impatiens. Table VI.— Showing the Effect of Slight and Excessive Drought on the Velocity of Ascent J, Distance Time in ' Velocity per x, in mm. ! seconds minute

Mean velocity = 70 mm. per min . 1 Mean velocity = 7-5 mm . per min . From the above tables we find that in the case of Chrysan- themum the velocity was decreased under excessive drought from the average value of 230 mm. to 18 mm. per minute, and in Impatiens from 70 mm. to 7*5 mm. per minute. These results prove conclusively that the velocity of ascent becomes decreased under increasing drought. I will next describe certain unexpected results which I obtained in the determination of the velocity. It is natural to expect that the velocity of ascent along the different flanks of the same stem would be the same. But in the determination of the velocity on the opposite sides of the stem of Chrysanthemum I found that though it was the same in a certain number of cases, it was widely different in others. Further investigation showed that the velocity was more or less uniform on all sides of specimens which had been grown in situations where the sunlight did not fall directly on the plant. In other specimens, of which the side facing south had been exposed to the action of sunlight, the north side being protected from it, though there was no visible difference in the two sides of the plant, yet an impressed physiological difference became revealed by the different speeds with which the sap ascended the two sides. I give in the following tables certain typical cases out of a large number.

Table VII. — -Showing the Difference in the Velocity of Ascent ON THE Sunny and the Shaded Sides of CJirvsanthemum It has thus been shown that, everything else being the same, (i) the velocity in a fully drooping specimen is, as previously shown, lower than in a semi-drooping specimen, and (2) the velocity of ascent in the shaded side is markedly higher than in the sunny side. This refers to the middle portion of the stem where the velocity is uniform.

This latter result also proves that osmotic action could not be the determining factor in the ascent of sap. For Arrhenius has shown that the osmotic pressure in a plant, is lower when growing in shade than in the open.^ The explanation of the lower velocity in the side stimulated by light has been arrived at by the experiments which will be described in detail in the next chapter ; it is that stimulus in general induces a diminution of velocity of ascent, and that this diminished velocity persists as an after-eftect of strong and long-continued stimulation.

In a plant subjected to drought, the root-cells cease to function from the absence of water-supply. The resultant diminution of turgor and hydrostatic pressure thus arrests the rhythmic activity underlying the ascent. If now, in a plant subjected to drought, water be applied to the top of the stem, it will be found that the direction of the flow of sap will be reversed, i.e. from above downwards. The following experiment demonstrates this in a striking manner. A drooping stem had all the leaves cut off except the terminal one. A beaker of water was raised so that the leaf was immersed in it (fig. 15). The result was that in a very short time the bent stem became erected, so that the leaf was lifted out of the water, the leaf having acted as an absorbent organ.

There next arises the question as to the relative rales of the flow of sap in the normal up direction and in the reversed down direction. This determination I have been able to carry out with great accuracy by the following arrangement (fig. 15). We make a vertical slit dividing the stem to a certain height into two halves, and place a piece of mica between the two ; the slit is carried above the right leaf c, and i cm. below the left leaf d. A specimen was chosen which had been uniformly exposed to the light from the sky and not to one-sided sunlight. The normal conducting power was thus the same on all sides. The lower end of the stem has a slanting cut, so that, by partly immersing the end, only the right half of the stem was supplied with water. Owing to the physio- logical interruption by interposition of the piece of mica, the movement of sap on irrigation was from A to B, and then to c, causing successive erection of the drooping