delivered the opinion of the court.
This is an action to recover compensation for damages occasioned by a fire charged against the railway company. There was a verdict for the sum of $60,000, which was reduced by the trial court to $42,500.
We shall use the terms plaintiff and defendant as they were there used.
The defendant is a trunk line railway, with double straight away tracks, running from Norfolk to Suffolk and beyond. Between these two cities is Dismal Swamp. That, as its name indicates, is a swamp. It is of vast extent, and stretches from Virginia into North Carolina. Originally it was covered by valuable forests, but they in a large measure have been cut and taken. This cut-over land has upon it a matted growth of bushes, rushes, reeds, cat tails and grass, which make it, in seasons of protracted drought, one great tinder box. The soil is in places covered with peat. It, too, burns when dry.
There are sixteen assignments of error, but when sifted down, in varying forms, they deal with these simple propositions : Did the fire spring from some passing train; and, if so, in the assessment of damages, have proper principles been applied?
*794The drought of 1930 was a memorable one. Certainly we have no record of one so complete or so disastrous. There were fires everywhere. In Norfolk county and in Nansemond county there were about three hundred. They came from passing trains, from hunters, from campers and from other causes. Some were promptly put out; others burned until the late winter rains. The State, railroads and private interests all fought them systematically and at heavy expense, but time and again the situation would get out of hand.
There is evidence for the plaintiff showing, or tending to show, that the fire which did the damage was set out by some passing train on October 4, 1930, and was first seen on the south edge of the right of way, near mile post 15. It began somewhere around 12:48 p. m., and if caused at all by a train, was caused by one of these:
The right of way to the south at this point was littered with weeds, grass, etc., cut but not burnt. The defendant probably thought that to burn it would be dangerous, and *795so as a substitute for that precaution, maintained small motor patrols which followed trains to see that fires were promptly extinguished. Because of some derangement in schedule, this patrol did not on this occasion follow until No. 99 had passed. The tracks run east and west. There was a stiff north wind blowing.
Robert Atkinson was chief fire warden for Norfolk and Nansemond counties. Under him was N. T. Poarch, also a fire warden and an employee of the Camp Manufacturing Company; under him were two negro patrolmen, Lester McCall and Ben Tann. They were stationed by him along the railway right of way, and their duty was to look for and put out fires. One of them carried a tank and the other a bucket. McCall said that he saw smoke for the first time just after eastbound freight train No. 89 had passed. It was about six telegraph poles away. He and Tann ran first to a water hole to fill their bucket and tank, and then to the fire, which was burning along the railway’s south ditch. They made two or three other trips to the water hole before Mr. Poarch came up. They were too late. He had been by there on patrol two or three times that morning but saw no fire to the north except a smouldering log from which no sparks were coming. On it they had poured water. This smouldering log was 130 feet from where the fire started, a little to the east, and thirty-seven steps from the railroad track. The water hole was on the north edge of the right of way and about a hundred yards from the fire.
On cross-examination he gave as an excuse for his failure the fact that he was cut off from this southside fire by a westbound freight, which must have passed about 2:25 p. m. He also said that he saw the fire ten or fifteen minutes after the eastbound freight had passed'. No. 89 passed at 12:48 going east and No. 99 at 2:25, so it is plain that he is mistaken in his estimates of time, but it is certain that at some period the water hole and one at it were cut off from the southside fire by passing freight train No. 99.
Ben Tann saw smoke on the south side just after this *796eastbound freight train had passed. He estimates that they were then about twelve telegraph poles from it. They went first to the water hole. When he reached the fire it was burning along the roadbed and leaping over the ditch. There was nothing on the north to burn. That section had been burnt over and there were left only two or three smouldering stumps or logs on which he had poured water. He further said that in the course of his patrol he had passed this point ten or fifteen minutes before and saw nothing. He made the same confusing statements about being cut off by a westbound freight train, and said on cross-examination that the fire had been burning but a short time when Mr. Poarch arrived.
Mr. Poarch was on duty that day, and had gone from his home near Franklin to fight fires if necessary. He posted McCall and Tann, after which he went east to about mile post 13. He tells us that there was nothing to burn north from where the fire started. After he had reached post 13, a coal train passed him going to Norfolk. Soon after it had passed a negro, Robert White, ■ came up and called his attention to the fact that there was a fire on the south side. He started towards a signal tower but was cut off by a westbound freight. When it had passed he saw from this tower smoke to which his attention had been directed. In a little while he followed this westbound freight train on a patrol motor car. When he reached the fire it was burning south and about 500 feet from the track. He found McCall and Tann carrying water and “wringing wet from perspiration.”
Benthall, one of the defendant’s witnesses, who was near mile post 16, first noticed smoke from this south side fire about half past one, while Whitehead, another of them, and who was between mile posts 17 and 16, thinks that it was about twenty or twenty-five minutes past one when he first saw it. We have seen that the eastbound freight train passed mile post 15 at 12:48.
On August 6th, a fire broke out north of the Norfolk and *797Western right of way, and burned continuously until some time in December. To protect the railroad and to limit the ravages of that fire, lands to the north and around mile post 15 had been twice burnt over by back fires, so that on October 4th there was nothing left in that immediate vicinity to burn. On that day the September fire was burning fiercely in what is known as the Gum Swamp, but that was somewhere from three-quarters of a mile to a mile and a half from this mile post and to the east. Also on October 4th, in this neighborhood, near the right of way, were two other fires, one known as the Moore fire and the other small and unnamed. On September 6th there were six other fires in this section, on or near the right of way.
Defendant’s engines were equipped with spark arresters. They are valuable preventive devices, but are not wholly efficient, as the many fires show. One of the plaintiff’s witnesses said all engines throw sparks. This is particularly true when they are pulling heavy trains. On the facts, as stated, plaintiff’s case rests.
It was and is the owner of three tracts of land—the Cook tract, of 2,429 acres; the McIntyre tract, of 1,299 acres; and the Big Entry tract, of 2,439 acres, making in all 6,168 acres, of which it claims that 5,500 acres were burnt over. All of this land is in one body, and runs with the defendant’s right of way, b.ut does not touch it at the point of origin of the fire. There intervenes a small lot owned by the Camp Manufacturing Company, across which the fire passed before coming to plaintiff’s holdings.
The facts as shown by defendant’s evidence, briefly stated, are: B. C. Smith, with Poarch as a passenger in his motor car, also went to the fire. He said that upon reaching it Tann undertook to explain what had happened. His explanation then was that wind-driven flames had leaped across the track. This Smith states was confirmed by fresh embers to the north which showed its progress. Other witnesses testified to the same effect. In other words, they say that the back-firing had not burnt the land clear, but *798had left patches of vegetation, one of them near mile post 15, and that from it the fire came.
A brakeman on train No. 1339, which passed at 11:50, said that as he passed mile post 15 there was a fire to the north about seventy-five feet away, and driven by a heavy wind.
An engineer of the westbound passenger train said that as he passed mile post 15 the wind was blowing hard from the north. Fire had then reached the right of way, with flames five or ten feet high, and in one place had crossed over. The fireman on train 89, which was the loaded eastbound freight, also testified to seeing a fire to the south which covered about a quarter of an acre. A brakeman on that train also saw it. Bradford, a hunter, said that he saw the fire coming from the north, driven by heavy winds, burn across to the south. The conductor on train 1339 said that as he passed mile post 15 bits of fire and burning leaves were falling on his cab, and that logs were blazing four or five yards to the north.
Without undertaking to go further into the defendant’s evidence, we may say that, if believed, it shows beyond peradventure this fire to be but a continuation of the old September fire to the north of its track; that on October 4th, sections of unburned vegetation at or near mile post 15 reached down to or near the right of way, and from its burning came all the trouble and not from any passing train. This is shown not only by eye-witnesses but by evidence of fresh burnings seen by witnesses who came later.
It brings us back to a situation which appellate courts are called upon time and again to face. We do not sit as a jury, nor are we a substitute for it. Here we have a verdict confirmed by the trial court. We must sustain it if it rests upon evidence which a jury could in reason have believed. Substantial conflicts in credible testimony are for it. This rule we have, in case after case, stated and restated.
In Virginian Railway Co. v. London, 148 Va. 699, 139 S. E. 328, 331, Judge Chichester said: “We conclude that the *799jury had a right to draw such conclusions as they did from all the circumstances in this case, and that this court has no right, whatever the weight of the evidence may have been, to set aside the verdict. Norfolk and Western Railway Co. v. T. W. Thayer Co., 137 Va. 294, 119 S. E. 107.”
We may sum up the plaintiff’s case: The right of way to the south was littered with dead grass and weeds. Land to the north had been burnt over, and there was on it at this time nothing inflammable—nothing which could supply material for a fire coming from that direction. All that was left was a smouldering, partly buried log, 130 feet away. Two witnesses tell us that the fire started shortly after a loaded freight train had passed. That train passed at 12:48. Another witness tells us that he was two miles to the east, and that soon after this freight train passed him his attention was called to smoke on the south. Two other witnesses who were near mile post 16 first noticed this smoke about a half hour after this eastbound freight had gone by. It further appears that about this time many other fires sprang up in this neighborhood along the right of way. If all this is true we cannot say that the verdict is unsupported. In passing, we may advert for a moment to Tann’s testimony. He said that he saw the fire soon after the loaded freight train had passed; that there was then no fire to the north except a few smouldering stumps on which he had poured water. He also denied that he had made any .statements to the effect that fire had leaped the right of way. There is evidence for the defendant to contradict him. If it be conceded that he did make them, that was a matter for the jury. Parsons v. Parker, post page 810, 170 S. E. 1, this day decided; Equitable Life Assur. Soc. v. Kitts’ Adm’r, 109 Va. 105, 63 S. E. 455.
If Tann and McCall were seeking to cover up their own negligence, they would scarcely have told us of their ineffectual efforts to put out a small fire. They might readily have charged their failure to the north wind and to a conflagration which no one could control.
*800It is true that the evidence is circumstantial. In cases of this kind it usually is. Proof of negligence is no longer necessary. When we have traced the fire to the railroad that is enough, made so by statute. Code, section 3992. Of course, it must be traced to the railroad, and this the plaintiff must do. It has that burden upon it which other plaintiffs usually have. It must prove its case. Proof that it has suffered loss is not sufficient. Southern R. Co. v. American Peanut Corp., 158 Va. 359, 164 S. E. 261. There a warehouse stood in small degree upon the right of way. A slow moving train had passed a half hour before the fire was discovered. When discovered it was “no bigger than a blaze from a lantern,” and just before the fire there had been a slight rain. Proof in that case was insufficient.
A fire itself may in certain circumstances be sufficient and in others not. Plainly the Chesapeake and Ohio Railway could not be held responsible for every fire which chanced to break out along its track as it passed through Richmond. But if a fire sprang up in an open prairie in dry grass in the wake of some train, we would have little doubt as to where it came from. Ordinarily inferences cannot rest upon presumptions but they may rest upon proven facts, and facts may be proven in many ways, all of which is but a roundabout way of saying that circumstantial evidence will do.
“Referring to the rule that an inference cannot be drawn from an inference or presumption, McCabe, J., in delivering the opinion of the court in Hinshaw v. State, 147 Ind. 334, at page 363, 47 N. E. 157, 166, says: ‘There is an important exception to that rule however. A fact in the nature of an inference may itself be taken as the basis of a new inference, whether intermediate or final, provided the first inference has the required basis of a proved fact. Burrill, Cir. Ev. p. 138; Best on Pres, section 187; 1 Greenl. Ev. section 34.’ (Italics supplied.) Reference to the authorities cited by Bailey on Personal Injuries, supra, and to the two Virginia cases cited, supra [Chesapeake & O. R. Co. v. Heath, *801103 Va. 64, 48 S. E. 508; Norfolk & W. R. Co. v. Cromer, 99 Va. 763, 40 S. E. 54], makes it clear that the presumption or inference referred to in the rule in question is a presumption or inference which may and often should be drawn but which, in the case in which the rule in question is applied is not a proved fact. If the inference or presumption is based on such evidence that it may be regarded as a fact proved in the case, then, although the evidence on which it is based may be circumstantial evidence, such inference or presumption of fact may itself form the basis for another inference of fact, equally as if it had been proved by direct evidence. 1 Wigmore on Ev. section 41.” C. & O. Ry. Co. v. Ware, 122 Va. 246, 95 S. E. 183, 186.
Usually a plaintiff should name the train which did the damage. That is not always necessary. They might pass a given point in such close succession as to make it unimportant. The railroad company and not the engine must answer. A fire may be traced to it by showing that it must have come from one or another of its locomotives. It might possibly occur on some double track line at the point of passing.
No new principle is stated in the Peanut Corporation Case. As we have heretofore stated, it lays down the familiar rule that a plaintiff must prove his case. Equally well established is the fact that he does not prove it merely by showing some loss by fire of property along the right of way. He must fortify it by evidence, direct or circumstantial. This is made plain in Virginian Ry. Co. v. London, 148 Va. 699, 139 S. E. 328; C. & O. Ry. Co. v. Ware, supra; Norfolk & Western Ry. Co. v. Spates, 122 Va. 69, 94 S. E. 195, and in many other cases. No one of these cases rests upon the naked fact that there was a fire. No court can undertake to say in advance just how much supporting evidence is necessary, and the evidence in no two cases is ever the same.
No plaintiff can rest his case upon mere proof of wrong; he must trace it to the defendant. He can do this by direct testimony or by a chain of circumstances, and this is the *802substance of what is held in every case which has come to us under section 3992 of the Code.
We think the evidence in this case is sufficient to support a verdict.
In the assessment of damages, have proper principles been applied?
As we have seen, there was a verdict for $60,000. The presiding judge thought that was excessive and reduced it to $42,500.
The average estimate of damage to the land, reproduction and merchantable timber as fixed by plaintiff’s witnesses is $70,242. The value of merchantable timber included in this estimate is $16,242. For the defendant it is said that this valuation is grossly excessive. If we had here only a conflict of evidence, the judgment of the jury should, of course, prevail. How was this verdict reached?
This instruction was given at the instance of the plaintiff, subject to the defendant’s objection:
D. “The court instructs the jury that if they find for the plaintiff, they shall assess plaintiff’s damages at an amount equal to the difference between the value of the property damaged or destroyed immediately before the fire and its value immediately after the fire, not to exceed the amount claimed in the declaration.
“If the jury believe from the evidence that the property damaged or destroyed, or any part thereof, had a market value at the time it was so damaged or destroyed, said market value shall govern in the assessment of damages. If, on the other hand, the jury believe that the property damaged or destroyed, or any part thereof, did not have a market value on or about October 4, 1930, then the jury shall proceed to arrive at the value of such property, or any part thereof, from a consideration of all the facts and circumstances proven.
“The court further instructs the jury that if they find that the plaintiff is entitled to recover damages, they may allow interest on the amount of damages resulting from *803the fire and fix the period from which said interest shall commence.
“* * * to the giving of which instruction the defendant objected and stated its objection to be as follows: That the proper measure of damage for standing timber trees is the difference in the market value thereof immediately before and immediately after the time said timber was damaged by the fire, and that the same measure of damage applies to the rest of the property of the plaintiff which was either damaged or destroyed by the fire; that 5,500 acres of land involved in this suit is alleged to have been damaged by the fire to the extent that a considerable portion of the soil thereof was burned away and destroyed.
“The defendant contends that the proper measure of damage for this land is the difference in the market value before and immediately after the alleged damage occurred and not the standard adopted by the court in Instruction D. The defendant contends that the Court of Appeals, in the case of Jeffress v. Virginia Ry. & Power Company, 127 Va. 694, 104 S. E. 393, has modified the doctrine of market value standard and has adopted money value, and that if the jury is of opinion there was no market value for plaintiff’s property immediately before and immediately after the fire, then the true measure would be the money value thereof.”
The issue here presented again appears in Instruction 7, offered on behalf of the defendant and refused.
7. “The court instructs the jury that should you be of the opinion from the evidence that the plaintiff has established the liability of the defendant for the fire, according to the instructions given on that subject, then the burden rests on the plaintiff to prove by a preponderance of the testimony the amount and extent of its injury, and, if you should decide that it is entitled to recover in this case, then the measure of its damage is the difference between the market value of its property burned by the fire immediately before and immediately after the fire, and this must appear *804from the evidence and the information obtained by the view.”
Plaintiff contends that all merchantable timber was destroyed or made valueless except a 168-acre tract of juniper, whose value was diminished by seventy-five per centum. It was not burnt over.
Most of the plaintiff’s witnesses say that the burnt over land, due to the burning, now has no value at all. One of them says that it is seriously and permanently injured, but still has some value.
The defendant does not contend that there was no damage done, but does contend that it has been exaggerated.
Where merchantable timber is injured or destroyed, we assess damages for such injury or destruction. If the growth is not merchantable, or is of special value, as a shade tree for example, we assess damages to the freehold. That is to say, the timber destroyed should be dealt with as such and the young growth on this land which had no merchantable value, but did add to its value, should be dealt with as a part of the land itself. Virginian Ry. Co. v. Hurt, 112 Va. 622, 72 S. E. 110, 112; Book v. Mammoth Copper Mining Co. (C. C.) 192 Fed. 748; Honaker Lumber Co. v. Kiser, 134 Va. 50, 113 S. E. 718; C. & O. Ry. Co. v. May, 120 Va. 790, 92 S. E. 801.
For timber destroyed, of course, this plaintiff is entitled to be paid whatever its value was at the time of destruction, but for timber not destroyed it is entitled to recover only such damages as it suffered—that is to say, the difference between the value of this timber just before the fire and its value just after the fire.
In Virginian Ry. Co. v. Hurt, supra, it was said that where the parties are sui juris, “the measure of damages in that case was the difference between the market value of the injured timber immediately before the fire and its market value immediately after the fire,” provided the timber had a market value.
In Chesapeake & Ohio Ry. Co. v. Baylor, 118 Va. 43, 86 *805S. E. 847, this instruction was approved: “The court instructs the jury that if they find for the plaintiff, then they shall assess his damages at an amount equal to the difference between the market value of the timber immediately before the fire, and its market value immediately after the fire.”
This timber, according to the plaintiff’s contention, was easily accessible. Its land was on the edge of Dismal Swamp. By it ran a trunk line railway, and by or near it a paved public highway and a navigable canal.
The measure or value for timber injured is its market value just before and just after the injury.
In the Hurt Case it was suggested that there might at times be no market value for injured timber. That distinction was not adverted to in the Baylor Case, which followed. Of course, where there is total destruction the case is simple, and it is correspondingly simple where there is no injury, and so no real controversy can come unless the timber continues in being as injured timber, in which case the ordinary rule for the ascertainment of damages is the rule stated. Here the situation from the plaintiff’s viewpoint is somewhat unusual. It claims damages for timber totally destroyed, and it claims damages for timber not injured but which was made less accessible by the fire and so less valuable.
It is true that courts are sometimes unable to apply the test of market value, but such cases ordinarily deal with property not brought into the course of trade and not known in the market. Judge Kelly, in Jeffress v. Virginia Ry. & Power Co., 127 Va. 694, 104 S. E. 393, 405, said: “Dwelling houses and shade trees apart from the land, and partly used or damaged personal property, though susceptible of valuation, are not usual articles of trade.” See, also, Murray v. Stanton, 99 Mass. 345.
We need not stop to consider this rule as applied to a house. In Norfolk & Western Ry. Co. v. Thomas, 110 Va. 622, 66 S. E. 817, Judge Harrison said that the measure of *806recovery is the value of property destroyed. In that case it appears that a dwelling was burnt.
These cases illustrate the exception to the general market value rule: Southern Express Co. v. Owens, 146 Ala. 412, 41 So. 752, 756, 8 L. R. A. (N. S.) 369, 119 Am. St. Rep. 41, 9 Ann. Cas. 1143. Here a book in manuscript was lost. The court recognized the rule, but said: “Where the article or thing is so unusual in its character that market value cannot be predicted of it, its value, or plaintiff’s damages, must be ascertained in some other rational way.”
Green v. Boston & Lowell R. Co., 128 Mass. 221, 35 Am. Rep. 370. A portrait was lost. The court said it had no market value.
Jones v. Morgan, 90 N. Y. 4, 43 Am. Rep. 131. It was there said that the market value test should not be applied to second hand furniture.
Mather v. American Express Co., 138 Mass. 55, 52 Am. Rep. 258. The court said the rule should not apply where architect’s plans had been lost.
In Sullivan v. Lear, 23 Fla. 463, 2 So. 846, 11 Am. St. Rep. 388, the court said that a franchise to use a wharf had no market value.
In McMahon v. City of Dubuque, 107 Ia. 62, 77 N. W. 517, 70 Am. St. Rep. 143, it was held that second hand wearing apparel and household goods had no market value.
An indefinite list of authorities might be cited to the same general effect. Long experience has demonstrated that compensation for articles which are the ordinary subject of commerce and trade, lost, destroyed or injured, should be measured by market value. Where such measure is manifestly inadequate, some other rule must be adopted. This is the substance of what was held in Chesapeake & Ohio Ry. Co. V. May, 120 Va. 790, 92 S. E. 801, 804.
When we come to deal with land, the situation is rather unusual. There is no case in this State in which damages for its destruction as such, have been asked. Here, as we have seen, plaintiff claims that this 5,500-acre tract, so far *807as value is concerned, has been totally destroyed. We have also seen that defendant concedes that it was damaged, but not that it has been made valueless.
The proper measure of damages for permanent injury is the diminution in market value.
Sedgwick on Damages (9th ed.) section 982, said: “The injury caused by the defendant may be of a permanent nature ; in such a case the measure of damages is the diminution in the market value of the (real) property.”
“The proper measure of damages for permanent injury to real property is the diminution in the market value of the property; * * *.” 8 R. C. L. section 44, page 481. In support of this proposition half a column of authorities are cited. See, also, the following late cases: Fuller v. Fair, 202 Ala. 430, 80 So. 814; Indiana Pipe Line Co. v. Christensen, 188 Ind. 400, 123 N. E. 789; Manning v. Woodlawn Cemetery Corp., 239 Mass. 5, 131 N. E. 287; Sebree v. Huntingdon Water Supply Co., 72 Pa. Super. Ct. 553; Houston & T. C. R. Co. v. Wright (Tex. Civ. App.) 195 S. W. 605; Gulf Pipe Line Co. v. Hurst (Tex. Civ. App.) 230 S. W. 1024; Thompson v. Illinois Cent. R. Co., 191 Iowa 35, 179 N. W. 191.
In C. & O. Ry. Co. v. May, supra, it appears that among other things destroyed by fire was a grove. That was worth something more than its timber value. The court said: “The evidence as to the loss of the grove was properly directed to prove by what amount its destruction diminished the market value of the property.” Its destruction did permanently injure it.
Of course, in some unusual situation this rule might not apply. A power company might flood a cemetery lot. Plainly it should pay, and its liability would not be affected by the fact that no market value could be shown.
In Feather River Lumber Co. v. United States (C. C. A.) 30 Fed. (2d) 642, 644, the court said: “As to the young growth, while the measure of damages in such a case is ordinarily the difference in the value of the land before and after the fire, here, there being no law to authorize the sale *808of the lands injured by the fire, the trial court admitted such evidence as was available to show the damage actually sustained, that is to say, what was required to make the government whole, * * *.” Since it was against the law to sell this land, it had no market value.
There is no magic in “market value.” The underlying purpose of the law is to make the wrongdoer pay for the damage he has done. In its assessment some general principle should be followed. It is a practical rule, has been found to work, and so has been adopted as a fair measure of value in all ordinary cases.
In the case in judgment plaintiff contends that its land was fertile, rich in agricultural possibilities; that it was in a large measure covered with young timber rapidly growing into value. Moreover, it is said that it was not in the heart of the swamp but on its edge, convenient, as we have seen, to navigable water, to a trunk line railroad and a paved highway. In such circumstances, it must have had market value. The situation in Norfolk county does not differ from that throughout the Commonwealth, and indeed from that throughout most of the States of the Union. It is a matter of common knowledge that for a time, ending in 1929, values were unduly inflated, after which the situation changed, and from then until now there has, compared with years gone before, been little demand for real estate, and the demand for other forms of property bought and sold in the ordinary course of commerce has correspondingly diminished. But it cannot be said that there has been no market. At the most, it can be said that there has been no market on the basis of those inflated values which had theretofore prevailed. To hold that there was no market would be to hold that there has been no market in Virginia in the past three years for land. They have been lean years, but realtors have lived.
As for timber, sawmills continue to operate, houses are being built, and lumber is bought and sold on the market every day.
*809Our conclusion is that there is nothing in the situation here to take this case out of the general rule, and that the trial court erred in refusing to apply it.
There are no other reversible errors in the record. Mr. Krimbill, a representative of the Federal Internal Revenue Department, testified that he examined this property in 1926 for the purpose of ascertaining its value as of March 1,1913. For reasons stated, his testimony should not have been admitted, but in a record so voluminous as this, errors of necessity sometimes creep in. When the case is heard again, this testimony should not be offered.
It is remanded for a new trial in accordance with the views here expressed, and on the single issue of damages.
Reversed and remanded.