Brannon, President:
Roman Pickens owned a water grist mill on Coal River in Kanawha county, and The Coal River Boom and Timber Company constructed a log boom on the river, below the mill, and Pickens claiming that the obstruction of the natural flow of water in the river by the boom works caused gravel, sand and mud to settle in the bed of the river instead of going on with the current, so much so that the sediment or deposit rose up to be even with the falls at the mill, sued the Boom Company for damage to the mill property: Pickens recovered a verdict and judgment, which were set aside in this Court. The report of* the case in 51 W. Va. 445 will state more fully the outline of the case. On a second trial Pickens obtained a verdict and'judgment for $5,000.00 damages, and the Boom Company brought the case to this Court.
Objection is made to instruction 1 of the'plaintiff because it told the jury that if the boom didin fact work damage to the mill, the finding should be for the plaintiff. It is argued that it denies the right of the defendant to locate and use its boom franchise in a reasonable manner, even though it was not chargeable with negligence in locating or operating the boom. “It is not lawful for one proprietor to impede or diminish the ordinary flow of water so as to materially interfere with the enjoyment of other proprietors.” Gould on Waters, section 218. “ Where there had been an injurious *13diminution in the flow of a stream to the use of which a mill owner is entitled, he may recover compensation for such loss. And where a mill owner is prevented from running his mill by an-interruption of the natural flow of a stream, he may recover the value of the use of such mill during the period of the interruption. And the determination of the rental value may be the measure of damages.” 3 Joyce on Damages, section 2146; 2 Famham on Waters, sections 546, 541. We shall not again discuss this law question. It is rm judicata from the effect of our former decision, that if the boom damaged the mill, Pickens had right to recover. The reason for so holding will be found in the two opinions on the former decision of this Court. I think that decision, in that result, is right; but it is foreclosed, right or wrong.
The objection to instructions 4 and 5 for plaintiff is that they give as the measure of damages, the rental value of the mill “ for the time ” the mill was stopped, whereas the opinion of Judge Dent said that the measure “is what the mill would have been worth during such deprivation of its use.” It is said that Pickens sought to heap up damages, and by the use of the word “for” instead of “during” mislead the jury to a fictitious rental value. We do not see that the point is substantial, the difference material, or the word misleading. It- would be straining the point to reverse a long trial for this cause.
Objection is made to plaintiff’s instruction 6. It told the jury that if the Boom Company secured a charter and a location for its boom, and made it large enough to catch logs, and afterwards the boom was leased to The Coal River Boom and Driving Company to operate for catching logs, and if the lessee company had used the boom, and additions which said lessee had made, and if the “location, building and operation of the said boom caused the injury to the plaintiff,” then the defendant could not escape liability on account of having made the lease. The argument • against the instruction is, that the lessee compans*' had added six cribs or piers to the eighteen existing at the date of the leasing- of the boom, and nothing in the lease warranted this increase, and that if the boom was not a nuisance when leased, the defendant was not liable, and that this instruction withdrew from the jury that question. It did not do so. It says “If the *14jury further find that the location, building and operation of the boom has caused the injury to the plaintiff,5 ’ thus demanding, as a test of liability?-, that the original location, building and operation of the boom caused the damage. Under this instruction damage merely from addition by the lessee would not make the lessor company liable. ‘ ‘ If the owner of lands through which the water course runs, erects a dam across it which lets the water back on the proprietor above, and then leases the land with the nuisance upon it, he gives with the lease implied permissson to the lessee to keep up the dam, and he thus becomes a participant with the lessee in the wrong while the dam is maintained as it was when he gave the tenant possession. ” Cooley on Torts, 725. If a landlord let premises already a nuisance, he and the lessee are both, or either, liable for the continuance. The landlord cannot shift the liability to other shoulders. And even if the lessee, by some work, add to the nuisance, but not by a separate, independent work, but one used along with the instrument of nuisance let to him, the lessor is still liable, though the injury come from both the work as it was when leased and the additional nuisance coming from the addition made by the lessee. The injury is the common fruit of the two, though lessor and lessee both contributed. How can you divide the injury — especiallj? in such a case as this ? Where the lessee creates, originates an independent, separate work, unauthorized by the lessor, the lessee is liable only. Where it becomes a nuisance only by the lessee’s act, and the landlord has not contributed, only the lessee is liable. 21 Am. & Eng. Ency. L., (2 Ed.) 721; 1 Jaggard on Torts, 225; 1 Kinkead on Torts, 99; 86 Am. St. R. 515-16-20. These principles are not denied; but it is said that the instruction fails to allow the jury to say whether the boom was a nuisance when leased. As shown above it makes it indispensable that the jury should find it to be, as originally located, the cause of injury.
Complaint is made that an instruction asked by the defendant was refused. It would have informed the jury that if the waterfall at the mill had been lessened by the boom, then “ unless the injury was done within five years next before the suit,” the jury could not find for the plaintiff. There are two answers to this complaint. This is the case of a *15continuing, but removable, nuisance. As held in the former decision, the plaintiff could recover for five years next before suit. It is not the case of a permanent, irremovable nuisance, doing injury, at its start, accomplishing its harm at once and for all time, so that limitation starts from the completion of the boom; but it is continuing, each day’s injury being a new wrong. The deposit of sand in the bed of the stream may have been, must have been, gradual, creeping slowlj' up the river and not harming the waterfall until long after the construction of the boom. So it was held in our former decision in this case, and this is res judi-cata. See Watts v. Norfolk R. Co., 39 W. Va. 196; Ells v. Chesapeake & O. R'y, 49 Id. 65. This instruction designs to tell the jury that limitation started with the completion of the boom, because it said that if the waterfall had been lessened “by the location, construction or operation of the boom,” then, unless that injury was within five years before suit, there could be no recovery. The instruction did not put the proper theory under the statute of limitations. The damage ensuing continuously recovery could be had as the damage came. That it was the purpose by the instruction to deny recovery if the building of the boom was more than five years before suit may be asserted, because the instruction totally denies any recovery on the hypothesis which it states; whereas, if such were not its import, it would have allowed recovery within five years. Secondly, if I misconstrue the instruction, and it puts the proper rule, then its refusal is not error, for the reason that instructions of the plaintiff clearly told the jury that there could be no recovery for damage done back of five years before suit.
Complaint is made that improper evidence touching assessment of damages was allowed. A witness was allowed to give rental value for the year 1892, and then to give it for each of five years before suit beginning April, 1894, and show the loss each year, diminishing the rental value yearly until it came to nothing the last year, owing to the growing harm from the yearly sediment. Then the questions were asked “What in your judgment, taking into consideration your knowledge of this mill property and its situation, would have been a fair rental value for it under the conditions existing in 1892, when you started to operate it for the following year ? ’ ’
*16“What in your judgment, based upon the same matters and things, would have been a fair rental value for that property under the conditions then existing between April, 1894, and April, 1895 ? ” and like questions as to subsequent years. It is asked by counsel, “ What bearing had the rental value of the mill for 1892 ? If the boom had never been built, is there any presumption that the rental value or earning value of the mill for 1892 would continue the same from year to year down to April, 1899 ? ” No presumption of law, and no presumption of fact, it may be true; and yet is it not a circumstance from which reasonable inference can be made? Now,the former decision held that the measure was the rental value during stoppage caused by the boom, and that the rental value should be the rental value of the property, “ as though the nuisance did not exist.” The year 1892 was just before 1893, in which injury to the mill began. It is right to take as a test or gauge some year before the boom was built. You have to do so. Why is not 1892 the fairest? Opinion of competent persons as to damages in such cases is proper. Miller v. Pulp Co., 30 W. Va. 567; Railroad v. Foreman, 24 Id. 663; Hargreave v. Kimberly, 26 Id. 787. How otherwise can you approximate? It is claimed that the number of days of stoppage ought to be shown. That is one process, but it is a demand with which it would be difficult to comply. There is a volume of evidence to prove the daily earning capacity of the mill, the quantity of grain which it grinds, the profit, etc. The evidence here objected to goes along with other evidence, and is not improper. The rental value was properly given in a year when the nuisance did not exist, and then for years during its existence. I think our former decision justified this.. It declared rental value the measure; that if the loss were total, then the loss of profits or rental value would be the measure. In case of movable nuisance, which.it held this to be, it is declared to be based on rental value. Witnesses acquainted with the mill, practical millers, were permitted to give rental value and the earning capacity daily. Why do not these constitute a basis for estimation of damages, in connection with evidence of the number of days of stoppage, which was given for some months?
Complaint is made that the evidence as to damage is specu*17lative, uncertain, furnishing no data or basis for measure. Watts v. Railroad Co., 39 W. Va. 210, is urged as authority for this ground of error. It holds that it would not be sufficient merely to show diminution of grinding power in the mill, but there must be some evidence to ascertain its extent, to show what was lost in earning capacity. In that case there was no evidence to do this; no evidence beyond guess or arbitrary speculation. We do not recant the principle as to this point stated in that case; but it is not governing in this case. It is wholly useless to go over the vast volume of evidence on the two trials bearing upon this matter. It consisted of opinion of earning power immediately before and after the erection of the boom given by Pickens, a practical miller, and millers who operated the mill for him, and Shanks, a practical miller at another mill in Putnam county; the quantity of different grains the mill would grind before and after the construction of the boom; evidence of profits thereon by practical millers; evidence by them of daily earning power in dollars; evidence for some months in different years of days lost by reason of inability to grind because of lessening of fall of water from the boom’s presence, which evidence was from memoranda made by the millers at the time; evidence of actual earnings by Pickens; evidence by practical millers of the earning capacity in dollars of the mill for each of five years before suit, diminishing as those years ran until the last of them, the last being estimated as having no earning capacity; evidence that this was chargeable to the boom; evidence not contradicted that the mill was worth, if not thus injured, $50,000; evidence afforded by a view of stream and mill by the jury on a view ordered. A volume of evidence under these general heads, and many, many facts and circumstances were before the jury upon what was a jury question, — the amount of damage. To say that no basis is furnished would require the impossible, and deny all recovery. It is impossible, in such a case, to fix amount with mathematical precision. “Absolute certainty is not required. Shall the injured party be allowed to recover no damage (or merely nominal) because he cannot show the exact amount with certainty, though he is ready to show, to the satisfaction of the jury, that he has suffered large damages by injury? Certainty, it is true, would be thus attained; but it would be *18the certainty of injustice. Juries are allowed to act upon probable and inferential, as well as direct and positive proof. And where, from the nature of the case, the amount of damages cannot be estimated with certainty, or only a part of them can be so estimated; we can see no objection to placing before the jury all the facts and circumstances of the case, having any tendency to show damages, or their probable amount, so as to enable them to make the most intelligible and probable estimate which the nature of the case will admit.” 1 Sedgwick, Damages, section 170. “It may be generally stated that even though the elements are uncertain and problematic from which the injury are to admeasure damages, nevertheless, they may be sufficient to constitute' the basis of a verdict. Therefore, the want of certainty is not of itself sufficient to warrant a refusal of damages.” 1 Joyce on Damages, section 75.
□ We do not here violate the rule against the allowance of damages merely speculative — mere guessed damages, merely visionary damages, resting on no other basis than hope, expectation, exaggerated idea of future profits, as laid down in Bodkin v. Arnold, 48 W. Va. 108, and James v. Adams, 8 Id. 568. Here we have certain data. Damages are given for the past on those data, not for the future by speculation or guess as to profits.
Excess of Damages. If the plaintiff is entitled to anything over nominal damages, we are safe in saying that the recovery is not excessive. The former trial resulted in the verdict of $12,190, this $5,000. But we shall not enter into details here. Evidence outlined above will repel the claim of excessiveness in amount of the verdict. Besides, that was a matter for the jury, and its verdict in case of damage to real property cannot be disturbed “unless plainly unwarranted by the evidence.” Miller v. Pulp Co., 38 W. Va. 558.
The only really grave feature of errors assigned in this case arises from the affidavits of four jurors filed upon the motion for a new trial, on which charge is made of misbehavior on the part of Pickens in treating the jurors to liquor at saloons during the trial. At the outset all the members of the Court desire to condemn with emphasis such an act. All the members of the Court think it not out of the way, but called for by the presentation, for the first time in the courts of the *19two Virginias, of this question, to express reprobation of this wrongful act. It is misbehavior constituting contempt. All members of the Court think that, if proven, it would set aside the verdict. Pour jurors say that Pickens treated them in saloons along with four other jurors. The affidavits aver specific occasions or instances of treating; but Pickens and the three other jurors said by three to have been present at the treating, deny any such treating on those occasions. And one of the impeaching jurors afterwards denies the treating on occasions specified by the other three, and says he never drank with Pickens privately. A deputy sheriff is said in the affidavits to impeach the verdict to have been present when the specified treating was done. He denies it. One juror says that he and another juror were going into a saloon as Pickens was coming out, and Pickens spoke, “ Set up the drinks to these boys,” or “ Boys go in and take something,” or words to that effect. This juror and others drank, but the juror says that another juror paid for the drinks, Pickens having gone out, not turning back into the saloon with them. Pickens swears that he never invited a juror to drink. Here is direct conflict, the preponderance against the charge of misbehavior. The burden is on the defendant to establish a charge at once condemnable and having the effect to nullify a long costly trial. There must be full proof of misconduct of a juror. 12 Ency. PI. & Prac. 559. “A finding by the lower court on conflicting evidence as to the misconduct of a juror will generally not be disturbed,” is the text of 17 Am. & Eng. Ency. L., (2 Ed.) 1209, on cases from many states. But suppose the affidavits of the jurors to impeach the verdict were not met by other affidavits. Are those affidavits admissible to impeach the verdict? In Bull’s Case, 14 Grat. 613, is a fully considered, deliberate discussion of the question whether jurors can be allowed to impeach their verdict, reviewing English and American cases, and laying down that, “As a general rule, the testimony of jurors is inadmissible to impeach their verdict; especially on the ground of their own misconduct. ” The distinguished Virginia jurist, Mon-cure, gives public policy as the reason for what at first blush seems an unreasonable rule, that to allow such evidence would be a strong inducement to a defeated litigant to tamper with, bribe or corrupt a juror to impeach with his solemn verdict; *20and it would enable a corrupt juror, really in favor of a defeated party, to give assent to a verdict, and then turn around and overthrow it. Cases where this rule may work wrong may occur; but the true public policy is to keep the door shut against the certain danger of the frustration and corruption of public justice from this source. No more dangerous door could be opened. That rule has been firmly followed in the Virginias. The opinion in Thompson's Case, 8 Grat. 650, says that the question is well settled in England against the competency of jurors to impeach their verdict, and that the great preponderance of American authority is the same way, and quotes with approval a Connecticut case holding that, “The opinion of almost the whole legal world is adverse to the reception of such testimony, and in my opinion on invincible foundations.” Harnsberger v. Kinney, 6 Grat. 287; Koiner v. Rankin, 11 Id. p. 431; Reed’s Case, 22 Id. 924; Bank v. Waddill, 31 Id. 483; Howard v. McCall, 21 Id. 212; Steptoe v. Flood, 31 Id. 323; Moses v. Cromwell, 78 Va. 671; Street v. Broadus, 96 Va. 823, are some of the cases stating this rule. In West Virginia this rule has been often followed. Vanmeter v. Kitzmiller, 5 W. Va. 380; Reynolds v. Tompkins, 23 Id. 229; State v. Cobbs, 40 Id. 718: C. & O. R. Co. v. Patton, 9 Id. 648; Bartlett v. Patton, 33 Id. 71; Graham v. Citizens Bank, 45 Id. 701. In Probst v. Breaunlich, 24 W. Va. 356, we find the rule, “It is settled in this state, as a general rule, with but few exceptions, if any, that the testimony of jurors will not be received to impeach their verdict.” In State v. Cartright, 20 W. Va. 32, it is held that evidence of jurors should be received only to support a verdict. So in State v. Harrison, 36 W. Va. 729, and in State v. Robinson, 20 Id. 713, it is held that such testimony may be used to support a verdict, but even then with great caution. So in Graham v. Citizens Bank, 45 W. Va. 701. If there were, but there is not, other evidence than that of jurors to prove treating by Pickens, we would be called on, by law, to grant a new trial. Where eating and drinking are furnished by the prevailing party it sets aside a verdict. Thompson's Case, page 657 of 8 Grat. I remark that scarcely a case can be found more fitly proving the wisdom of the rule rejecting the evidence of jurors to impeach their verdict than this case. Three or four jurors swear to the treating; four *21deny the same treating. Iiave those on one side been tampered with ? Which ones? Strange, it seems, that if Pickens was so liberal in treating .that others, bartenders or bjr-standers, (some are generally present) are not called on to prove the treating. As the burden is on the defendant, and especially as counsel knew that the evidence of jurors was at least questionable, we would expect that outside testimony would be produced. Does not the law demand it?
An array of cases from other states is cited to support the affidavits of jurors to impeach their verdict. We shall not review them here, as it would be a prolix work, and useless, because in the face of so many Virginia and West Virginia cases, through at least fifty years, we cannot follow another rule. The foreign cases are cited to sustain an exception to the rule sought to be made in this case based on the theory that affidavits of jurors may be taken to prove misbehavior outside the jury room, and may be received to prove misbehavior of a parti/ to the suit. The answer to this, as applied to this case is that the same thing which would be misconduct or misbehavior in Pickens is, if possible, worse misbehavior an the part of the jurors treated. If they drank Pick-ens’ whiskey, they knew it was furnished to induce them to favor him, to bribe and seduce them, and it was gross misconduct in them; and the law brands it as misconduct in them, and will not open its ear to hear them impeach their verdict by their own misconduct. The motion for a new trial rests in fact on the misconduct of some jurors. There is no claim or evidence that any juror drank to excess, or that any one was intoxicated.
Judgment affirmed.
Affirmed.