dissenting.
I must respectfully dissent from the af-firmance of the judgment entered on the jury verdict in favor of Elizabeth Morris. The majority determines that the trial court could reject an inference that the jury used a quotient verdict because of the varied quotients on the different jury memoranda collected from the jury room. I disagree with the view that the trial court is permitted to weigh the multiple inferences that may arise from the memo-randa from the jury room in order to determine whether that memoranda raises the presumption that the jury employed the quotient process. My understanding of the law governing quotient verdicts differs from that of the majority because I believe that if any fair inference, even if it is but one of many possible inferences, indicates that the jury used the quotient *344process, the presumption that the jury used the quotient process arises. In order to better explain my basis for dissenting, I will set out my understanding of the applicable law in detail.
A presumption of a quotient verdict may arise “where data found in the jury room, and appearing to be the work of the jury, produces a quotient substantially the same as the verdict rendered.” Security Mut. Fin. Corp. v. Harris, 288 Ala. 369, 372, 261 So.2d 43, 45-46 (1972). A jury may “use the quotient process for the purpose of obtaining a figure that represents an average of the amounts the individual jurors feel should be awarded and that serves merely as a suggestion or basis for further discussion, deliberation, or consideration.” Warner v. Elliot, 573 So.2d 275, 277 (Ala. 1990). However, the fact that a verdict is substantially equal to or approximates the quotient reached will result in a presumption that the jury impermissibly relied upon the quotient process in order to return a verdict. See Harris, 288 Ala. at 373, 261 So.2d at 46. As our supreme court explained in Hams:
“In this state, it has become established that where evidence is presented of scraps of paper, lists of figures and other memoranda, indicating that the jury used the quotient process and obtained a quotient which corresponds with the amount of the verdict or approximates it, a presumption arises and a prima facie case is made that the jurors have improperly used the quotient process in connection with an antecedent agreement to be bound by the amount of the quotient and that they have thus rendered an invalid quotient verdict....
“ ‘ ... It is a settled rule in this state that when figures are shown which were used by the jury in its deliberations and from these figures a fair inference may be drawn that the verdict was a quotient verdict, the court will so hold and that the verdict was the result of a previous agreement, unless the contrary is shown.’ ”
Id. (quoting Fortson v. Hester, 252 Ala. 143, 147, 39 So.2d 649, 651 (1949)). Put more simply,
“when the record shows scraps and bits of paper containing numbers, lists of figures and computations indicating the use of the quotient process to arrive at an amount to be awarded to a party to the lawsuit and the amount awarded closely corresponds to the amount obtained by that process, a presumption arises that the quotient process has been used in conjunction with a prior agreement to be bound by the results of said process.”
Alabama Power Co. v. Thomas, 50 Ala. App. 517, 519, 280 So.2d 778, 780 (Civ. 1973). Notably, variances between the quotient reached and the verdict awarded do not negate the presumption that the verdict is an impermissible quotient verdict. Harris, 288 Ala. at 372, 261 So.2d at 45 (“[Pjrecise agreement between the quotient found and the verdict returned[] is not required.... [A] quotient verdict is invalid even if the amount of the verdict is not exactly the same as the quotient obtained by the jury, but is reached by rounding off the quotient to an even number or by making some other slight addition or subtraction.”).
The basis for allowing a presumption of a quotient verdict to be created is rooted in the fact that, in Alabama, juror affidavits may not be permitted to impeach a verdict but may be used to uphold a verdict. See, generally, Alabama Power Co. v. Brooks, 479 So.2d 1169, 1178 (Ala.1985) (“Neither testimony nor affidavits of jurors are admissible to impeach their verdicts; however, such evidence is admissi*345ble to sustain them.”). Our supreme court has explained its rationale this way:
“Courts have much regard for the verdicts of juries, and are indisposed to presumptions tending to overturn them. But, in view of the case with which the winning party may produce explanatory evidence in cases of this kind, and the inability of the losing party to obtain other than circumstantial evidence, we think the rule [permitting the presumption of a quotient verdict] convenient of application and conservative of justice.”
George’s Rest v. Dukes, 216 Ala. 239, 241, 113 So. 53, 54 (1927). It has long been the law of this state that the party challenging a verdict on the ground that it is a quotient verdict may not use affidavits of jurors to establish that, in fact, the jury used the quotient process. Birmingham Ry. Light & Power Co. v. Moore, 148 Ala. 115, 130, 42 So. 1024, 1030 (1906), overruled on other grounds by Birmingham Ry. Light & Power Co. v. Goldstein, 181 Ala. 517, 532, 61 So. 281, 283 (1913). Thus, the challenger is entitled to a presumption that the verdict was arrived at through a quotient process when the notes of the jury containing the figures it used to calculate the verdict amount provide a basis for a fair inference of the use of such process. Harris, 288 Ala. at 373, 261 So.2d at 46; Fortson, 252 Ala. at 147, 39 So.2d at 651. The presumption may then be rebutted by affidavits from the jurors refuting the inference that a quotient process was used. See City of Dothan v. Hardy, 237 Ala. 603, 608, 188 So. 264, 268 (1939) (affirming the trial court’s denial of a motion for new trial when the evidence included affidavits from jurors indicating that the jurors had abandoned the use of the quotient process); Birmingham Ry., Light & Power Co. v. Clemons, 142 Ala. 160, 162, 37 So. 925, 925 (1904) (“It was competent for the plaintiff to prove by the jurors themselves in support of their verdict thus sought to be impugned, that these figures were made and that this process was resorted to without previous agreement that the result should be the verdict....”).
In support of his motion for a new trial, Matthew Morris presented several pieces of paper recovered from the jury room. Upon one piece of paper, written in columns, appear the names of the 11 jurors10 and various numbers ranging from 25 to 175; at the top of that piece of paper is written $115 (“jury note 1”). On a second piece of paper is written “100 punitives,” which is crossed out, followed on separate lines by “5,000 punitive FI,” “Comp A,” “100,” “115K,” and “$220” (“jury note 2”). The third piece of paper contains the figure “75/125.00” (“jury note 3”). On a fourth piece of paper appears a column of 11 numbers containing four “150” entries and seven “220” entries (“jury note 4”). On that same piece of paper appears “$220,” “$215.00,” which was crossed out, “$195,” and “$200.00.” When the numbers in the column on jury note 4 are added together and divided by 11, the quotient is $194.54. A fifth piece of paper contains a column containing the numbers “$25,000,” “$25,000,” “$29,000,” which is crossed out, “$75,000,” which is also crossed out, and “$100,000” (“jury note 5”); in a second column, jury note 5 contains the words “loss of time,” “injury,” “restrained of freedom,” “assault,” “medical bills,” “fleed [sic] from scene,” and “time loss of work.” The amounts appear to be totaled below the aforementioned figures on jury note 5; the original total was “$150,000” with the number “$175,000” ’ written over the original *346total. Other computations appear on jury note 5, but they appear irrelevant.
Although Matthew focuses on the figures on jury note 1 described above, I find the figures on jury note 4, and specifically the quotient reached by totaling those numbers, more compelling. I agree with Matthew that the figures on jury note 4 are shorthand for $150,000 and $220,000; there is simply no other explanation for the numbers $150 and $220 when the verdict ultimately rendered was $200,000. The quotient of the sums on jury note 4 is $194.54; thus, the quotient is shorthand for $194,540. The jury awarded Elizabeth $200,000.
I believe that to determine whether the jury notes, which both parties admit were the work of the jury, raise the presumption that the verdict was a result of the quotient process a court must simply ask if the notes raise, as one possible inference, the concern that the quotient process was used. Notably, the terminology used by our supreme court in Forbson, 252 Ala. at 147, 39 So.2d at 651, does not require that the jury notes or other memoranda prove or establish the use of the quotient process. The jury notes or other memoranda must merely raise “a fair inference” that the quotient process was used by the jury.
“An inference is a deduction of fact that reasonably may be drawn from another fact or group of facts. It is in the very nature of inductive, as opposed to deductive, reasoning that the same premise, or set of premises, will give rise to more than one inference, each of which is logically compatible with the initial premise or premises. Merely because the same factual premise will support more than one inference does not of itself render the proffered evidence conjectural or speculative. Indeed, inferences may be of greater or lesser persuasion even though, as a matter of strict logic, they may all follow rationally from the same premise.”
Roberts v. Carroll, 377 So.2d 944, 947 (Ala. 1979).
Based on the language employed by our supreme court in Fortson, the party challenging the verdict on the ground that it is a quotient verdict must present jury notes or memoranda that would support a fair inference that the jury used the quotient process. The presumption then arises that the jury agreed in advance to be bound by a quotient process, and the trial court must then “so hold” that the verdict is a quotient verdict, subject to rebuttal by evidence presented in support of the verdict. See Warner, 573 So.2d at 277; Harris, 288 Ala. at 373, 261 So.2d at 46. The jury notes or memoranda, provided that they would support a fair inference that the quotient process was used, themselves trigger the presumption. Warner, 573 So.2d at 277. In Warner, our supreme court stated that “the presumption of a quotient verdict ... was created when [the challenger of the verdict] introduced the scraps of paper from the jury room[, which contained a list of figures added together and divided and a quotient approximately the same as the verdict awarded,] into evidence.” Id. Similarly, in Harris, our supreme court stated that the presumption that the verdict was a quotient verdict arose after the challenger of the verdict presented “incriminating papers” containing a quotient corresponding substantially to the verdict and after those papers were “adequately identified” as having been created by the jurors. Harris, 288 Ala. at 373, 261 So.2d at 46. As noted above, our supreme court has explained that the reason for the presumption is rooted in the fact that the challenger of the verdict has the opportunity to present only circumstantial evidence of the use of the quotient process while the proponent of the verdict *347may present explanatory evidence with relative ease. Dukes, 216 Ala. at 241, 113 So. at 54; Harris, 288 Ala. at 873, 261 So.2d at 47. Thus, the burden on the challenger of the verdict is not to prove by a particular quantum of evidence that the jury notes or memoranda were used by the jury in executing the quotient process.11 Instead, all that the jury notes or memo-randa must do is provide a basis for a fair inference that they were so used. See, e.g., Fleming v. Knowles, 272 Ala. 271, 278, 130 So.2d 326, 332 (1961) (stating of a jury note containing a list of 12 figures and a quotient approximating the verdict: “[djata of this kind here shown have been held prima facie sufficient to impeach the verdict as a quotient verdict”); Hardy, 237 Ala. at 608,188 So. at 268 (same).
Of course, in determining whether the jury notes or memoranda give rise to a fair inference that the quotient process was used, the trial court, and this court on appellate review, must consider whether the quotient on the jury notes or memo-randa “corresponds with” or “approximates” the verdict. Alabama law provides that a slight difference between the quotient and the jury’s award will not affect the presumption that the verdict was decided by the quotient process. In Harris, our supreme court held that a $42 difference between the quotient and the verdict awarded was not significant and could be explained as the jury’s decision “merely [to] round[] off [the award] to the next hundred.” Harris, 288 Ala. at 373, 261 So.2d at 46. Our supreme court has also indicated that a larger discrepancy may still give rise to the presumption, indicating in Hardy, 237 Ala. at 608, 188 So. at 268, that the presumption of a quotient verdict arose when the quotient was $6,700 and the jury awarded $6,000. In Maring-Crawford Motor Co. v. Smith, 285 Ala. 477, 488, 233 So.2d 484, 494 (1970) (opinion extended on rehearing), our supreme court held that a fair inference that the verdict was a quotient verdict arose where jury notes indicated that the jury had added together 12 figures and had divided the sum by 12 to reach $16,125 and the verdict awarded was $15,000. More recently, the supreme court determined that the presumption of a quotient verdict was raised when the jury notes recovered from the jury room yielded a quotient of $34,670 and the verdict returned by the jury was $30,000. Warner, 573 So.2d at 277. However, this court has held that a $225 variance between the quotient — $3,275—and the verdict — $3,500—was “much more than a mere rounding off or a slight addition or subtraction to the quotient amount” and determined that the difference, which was more than the $42 amount permitted *348in Harris, negated any conclusion that the verdict was a quotient verdict. Burgreen Contracting Co. v. Goodman, 55 Ala.App. 209, 222, 314 So.2d 284, 296 (Civ.1975). Based on consideration of precedent other than Harris, I conclude that Goodman does not comport with the holdings of our supreme court regarding quotient verdicts.
Elizabeth argues in her application for rehearing that the difference between the actual quotient of $194,540 and the jury’s verdict of $200,000 is too large to be considered a slight deviation and, therefore, that the presumption that the jury used the quotient process did not arise. I cannot agree with Elizabeth that our focus should be limited to the numerical difference between the quotient and the verdict. As our supreme court stated in 1913, “[t]he evil effects of a quotient verdict cannot be cured by agreeing thereafter to a slightly different verdict, if it appears that the agreement made in advance entered into or induced the result....” International Agrie. Corp. v. Abercrombie, 184 Ala. 244, 259, 63 So. 549,'553 (1913).
The Abercrombie court went on ‘to discuss quotient-verdict cases from other states, including one from Texas, from which the court quoted the following test:
“ ‘ “The test in such cases is: Did the jury agree to be bound beforehand by the result of such proceedings? If so, the verdict will be set aside. But if the result was reached, and no agreement had beforehand to abide by it, and it was afterwards agreed upon as their verdict, it will be sustained.” ’ ”
Abercrombie, 184 Ala. at 259-60, 63 So. at 554 (quoting Texas Midland R.R. v. Ather-ton, 123 S.W. 704, 704 (Tex.Civ.App.1909), quoting in turn Missouri, Kansas & Texas Ry. Co. of Texas v. Hawkins, 50 Tex.Civ. App. 128, 133, 109 S.W. 221, 224 (1908)).
The Abercrombie court also relied on a Kansas case in which the jurors, after determining the quotient, which was “nearly $300,” decided to make the verdict an even amount and set the verdict at $300. Abercrombie, 184 Ala. at 261, 63 So. at 554 (citing Ottawa v. Gilliland, 63 Kan. 165, 65 P. 252 (1901)). The Abercrombie court then quoted the following portion from the opinion in Ottawa:
“ ‘After the amount was found by marking, aggregating, and dividing there was no reconsideration. The addition to that' amount was not made after a further consideration and the decision of the cause upon its merits, but was for the one purpose of making an even amount. The law demands of each juror an honest consideration of the rights of the parties litigant and the exercise of his best judgment, guided by the law and the evidence of the case. A verdict reached in any other way should 'be set aside.’ ”
Abercrombie, 184 Ala. at 261, 63 So. at 554 (quoting Ottawa, 63 Kan. at 166-67, 65 P. at 253).
The present case is complicated by the fact that the jurors used shorthand numbers to figure their verdict. Clearly, the “rounding up” from $194.54 to $200.00 would not be considered a substantial deviation such that the presumption that the verdict was a quotient verdict would be overcome. When one considers that the numbers actually stand for $194,540 and $200,000, it is much more difficult to say that the difference between the quotient and the verdict is not substantial in a numerical or monetary sense. However, our supreme court has announced that a presumption that a quotient verdict exists is to be drawn when the figures used by the jury in its computations give rise to “a fair inference” that the jury determined the amount of their verdict by use of a quotient process. Fortson, 252 Ala. at 147, 39 So.2d at 651. Thus, I believe that the *349trial court, and this court on appeal, must determine what “fair inference” arises from the figures on jury note 4. In doing so, I conclude that the focus should not be on the numerical difference between the quotient and the verdict but, instead, should be centered on whether the evidence gives rise to the fair inference that the jury’s verdict was induced by the use of the quotient process. See Abercrombie, 184 Ala. at 259, 63 So. at 554 (“The evil effects of a quotient verdict cannot be cured by agreeing thereafter to a slightly different verdict, if it appears that the agreement made in advance entered into or induced the result.”); see also Benjamin v. Helena Light & Ry. Co., 79 Mont. 144, 255 P. 20, 24 (1927) (noting the lack of a “well-defined rule about what discrepancy is slight nor how much the discrepancy may be and the verdict still be invalid” and concluding that “[mjuch seems to rest on the basic facts that mathematical calculation was resorted to and led up to the fixing of the amount of the verdict, even though not the same as the quotient”). Otherwise, I believe that we run the risk of allowing a decision by the jury to increase or decrease the quotient by more than a minimal sum to “cure the evil effects of a quotient verdict,” which such decisions have long been held unable to cure. Ledbetter v. State, 17 Ala.App. 417, 418, 85 So. 581, 582 (1920); see also Aber-crombie, 184 Ala. at 259, 63 So. at 554; and Stone v. State, 24 Ala.App. 400, 400, 135 So. 646, 647 (1931). Such a conclusion also comports with the reason behind permitting the presumption to arise in the first place — the comparative ease with which the proponent of the verdict may produce evidence of the jurors’ deliberation process to support the verdict.
The numbers on jury note 4 lead to the fair inference that the jury used the quotient process to come up with the number “$195,” which is merely $.46 more than the $194.54 arrived at by the quotient process. On the same note, written in close proximity to “$195,” is “$200.00.” A fair inference can be drawn that the jury concluded that it should round $195 to $200, an even number. Jury note 3 indicates that the jury then concluded that it should divide the $200 damages award into “75” in compensatory damages and “125.00” in punitive damages.
I cannot agree that the $5,460 difference between the actual award of $200,000 and the quotient of $194,540 negates the inference that the jury used the quotient process to reach its verdict based on its use of the shorthand numbers $195 and $200. Although that difference is larger than any other discrepancy between a quotient and a verdict mentioned in our caselaw, I note that in Warner the $4,670 difference between the verdict and the quotient involved a $30,000 verdict, as opposed to a $200,000 verdict. Percentage-wise, the difference in the present case is smaller than the amount recognized to trigger the presumption of a quotient verdict in Warner. Thus, I cannot agree that the inference raised by jury note 4 was negated by the apparent decision to increase the verdict to a round number more easily capable of division, because Elizabeth provided no proof that such adjustment was made “ ‘after a further consideration and the decision of the cause upon its merits.’ ” Aber-crombie, 184 Ala. at 261, 63 So. at 554 (quoting Ottawa, 63 Kan. at 166-67, 65 P. at 253); see also Warner, 573 So.2d at 277 (“If, after obtaining the quotient, the jurors continue to deliberate and thereafter arrive at a different amount of damages, the verdict will not be considered a quotient verdict and will be sustained, absent some other infirmity.”).
At oral argument, a suggestion was made that the fact that the jury was polled and each juror indicated that the verdict *350was his or her own should negate the inference that the verdict was a quotient verdict. Our supreme court has stated that a verdict induced by the quotient process cannot “be cured by the jury subsequently adopting it, or its equivalent, as their verdict, if the agreement entered into or controlled the subsequent adoption of the verdict returned.” Abercrombie, 184 Ala. at 259, 68 So. at 558. Thus, the polling of the jurors did not serve to negate the presumption raised by jury note 4.
Based on the above-examined authorities, I conclude that Matthew produced sufficient evidence from which arose a fair inference that a quotient process was used by the jury. The jury’s verdict of $200,000 closely approximates and is substantially similar to the quotient contained on jury note 4. Thus, I conclude that he was, and is, entitled to the presumption that the jury’s verdict was an impermissible quotient verdict.
As noted above, Elizabeth was permitted to offer affidavits from one or more jurors to rebut the presumption that the verdict was a quotient verdict. See Warner, 573 So.2d at 277 (affirming the trial court’s exclusion of juror affidavits that tended to impeach the verdict while noting that the testimony of the jury foreman, which rebutted the presumption of a quotient verdict, was properly admitted). Our supreme court has explained:
“Evidence of jurors on a motion for a new trial is admissible to sustain their verdict. It may be shown the memoran-da, if made by the jury, was a mere tentative expression of the jurors’ views for the purpose of further deliberations; that no prior agreement, express or implied, to abide the result, was entered into; that such suggestion, if made, was abandoned, or other facts to show the verdict was the expression of the fair judgment of the several jurors.”
Dukes, 216 Ala. at 240, 113 So. at 54. However, Elizabeth did not present any evidence to rebut the presumption arising from jury note 4. See Harris, 288 Ala. at 373, 261 So.2d at 46 (“The appellee produced no testimony or affidavits to prove that the jurors did not agree in advance to abide by the result of their computations, or that the quotient was only a basis for further deliberations, or any other facts to show the verdict was an expression of the fair judgment of the several jurors.”); Dukes, 216 Ala. at 241, 113 So. at 54 (“In this case no evidence from jurors or other source was offered to question the connection of these papers with the verdict, or to otherwise explain their implication.”). Without such evidence, the trial court had no basis for rejecting the presumption that the verdict was an impermissible quotient verdict. Accordingly, I dissent from the majority’s opinion.
. The record indicates that 1 of the jurors was excused and that only 11 jurors deliberated.
. The term "fair inference” is used more often in Alabama’s criminal jurisprudence in cases in which the appellate court is reviewing a trial court’s decision to overrule "a motion to exclude the state's evidence” and to enter a judgment of acquittal. See Thomas v. State, 363 So.2d 1020, 1022 (Ala.Crim.App. 1978). The standard of review of the trial court’s decision on such a motion is “whether there exists legal evidence before the jury, at the time the motion was made, from which the jury could by fair inference find the defendant guilty.” Thomas, 363 So.2d at 1022. The Thomas court further explained that the jury alone has the power to determine the "probative force and weight of the evidence” and that an appellate court may only determine whether the evidence presented could support a fair inference that would support a guilty verdict. Id. That is, the appellate court looks to determine whether a jury could fairly infer that the defendant was guilty based on the evidence that was the subject of the motion to exclude and not whether the trial court or the appellate court would so infer. Id. This supports the conclusion that the requirement that the jury notes or memoranda create a fair inference that the quotient process was used requires only that such an inference could be drawn, as one of several possible inferences, before the presumption that the verdict is a quotient verdict is raised.