delivered the opinion of the court:
Plaintiffs, Jack Gilbert and Bill Wade, appeal from a judgment of the trial court for defendant, Old Ben Coal Corp., following a bench trial of their complaint in which they alleged violations of the maximum hour and overtime pay provisions of the Fair Labor Standards Act of 1938 as amended (hereinafter referred to as the Act) (29 U.S.C. §201 et seq. (1976)). Pursuant to section 2071 of the Act, they each sought unpaid overtime compensation, earned in the capacity of “resident mining engineer,” as well as liquidated damages in an amount equal to the alleged unpaid overtime compensation and attorney’s fees.
*489Each plaintiff alleged in his complaint that he had worked “a 50 hour work week” during the pertinent period of time, which was for plaintiff Gilbert the two years immediately prior to the filing of the complaint on February 20, 1976, and for plaintiff Wade the period between February 20, 1974, and September 11, 1975, when defendant terminated his employment. Defendant claimed in defense that by virtue of plaintiffs’ “executive, administrative or professional capacity” plaintiffs were, under the provisions of section 2132 of the Act, exempt from the operation of the overtime provisions of section 207.
The trial court found that defendant had sufficiently proved that during the period of time involved plaintiffs were “exempt employees employed in a bona fide administrative capacity.” The court found further that plaintiffs “failed to prove that the employees of the business in question were engaged in interstate commerce and the amount of overtime involved.” The court observed that “[t]he plaintiffs were impeached in relevant particulars. We have only their statements, without more, that they worked an average of 50 hours per week.” Defendant had kept no records pertaining to the hours plaintiffs worked. With regard to that omission of defendant, the trial court cited the case of Brennan v. Parnham (W.D. Pa. 1973), 366 F. Supp. 1014, where in the absence of any records of the employer pertaining to the amount of time the employees had worked, the court determined the number of overtime hours worked by the employees, in part, at least, from the “reasonable and creditable estimates of the employees themselves.” The instant court found, by comparison, that “the record is so devoid of factual evidence that the court has no basis for deciding that the employees’ estimates were ‘reasonable and creditable.’ A lay witness who gives an opinion, if we may so characterize these conclusions as such, must give the basis of his observations as a foundation for such opinion. If evidence is admitted without objection which is basically without sufficient foundation the court may not assume that the opinion is reasonable if that is a necessary part of the required proof.” The trial court ordered that plaintiffs recover nothing and that defendant recover costs of suit from plaintiffs.
From a denial of both plaintiffs’ post-trial motions to vacate the judgment for defendant and to enter judgment for plaintiffs or, in the *490alternative, for a new trial, plaintiffs appeal, presenting a number of issues for review. Essentially plaintiffs contend on appeal that the trial court should have granted their motions because of the sufficiency of plaintiffs’ evidence and that in view of plaintiffs’ evidence the court erred in its findings that (1) defendant proved plaintiffs were exempt employees in a bona fide administrative capacity, (2) plaintiffs failed to prove defendant’s employees were engaged in interstate commerce, and (3) plaintiffs failed to prove the amount of overtime involved.
The law is well settled that the trial court, having had the opportunity to see and hear the witnesses, is entitled to consider the evidence and that the reviewing court is under a duty to reverse any judgment “wherein the findings are clearly and palpably against the manifest weight of the evidence [citations], or wherein there is a clear and palpable error in some other respect.” (Drovers National Bank v. Great Southwest Fire Insurance Co. (1977), 55 Ill. App. 3d 953, 956, 371 N.E.2d 855, 857.) We have examined the record on appeal and have determined that the trial court’s finding with regard to plaintiffs’ failure to prove the amount of overtime involved is not contrary to the manifest weight of the evidence. Therefore the trial court did not err in denying plaintiffs’ motions for a new trial.
Since the basis of plaintiffs’ claim is that defendant owes them compensation yet unpaid for their labor in excess of 40 hours per week over the course of many weeks, our affirmance of the trial court’s finding with respect to the amount of overtime renders unnecessary any consideration by us of the other issues plaintiffs raise. Consequently, we cite only those facts which bear on the issue of the amount of overtime plaintiffs worked.
Though plaintiffs did not work in the same mine, the testimony of one mirrored that of the other in the description each gave of his work as a “resident mining engineer,” or, as the position is also called, “mine surveyor.” According to their testimony, plaintiffs were responsible for setting, by the use of surveying tools and principles, numerous underground “sites,” which the miners use as a point of reference primarily in the operation of the continuous mining machines. These “sites,” which indicate the center of the path a mining machine should follow, are the means by which the operators of the machines keep them on a straight, safe course. Working from a federally approved mining plan, plaintiffs, each with the aid of a person known as a “helper,” were responsible for placing the “sites” so that the actual mining conformed to the mining plan, to the extent that it was possible for it to do so, given inevitable obstructions to the mining such as the discovery of water in the mine. The “sites” each consist of a pair of wooden plugs, which the helper drives into the roof of a room of the mine at a location behind the mining machine. From the plugs, which are set, if possible, three to four feet *491apart, the miners hang plumb bobs, which they observe as they operate the mining machines. Under good mining conditions a single pair of “site” plugs might serve as a guide for the mining of 200 feet of coal. The progress of the mining machines varies as a result of more or less difficult mining conditions. As the mining progressed, plaintiffs were required to set new “sites.” In the mines in which plaintiffs worked, a single resident mining engineer advanced the “sites” for all three shifts of approximately 100 miners per shift. Miners who worked the night shift, for example, might well have use<^ “sites” plaintiffs had set during the day shift. Plaintiffs were permitted to choose the time of day when they would work and were free to leave the mines when they had completed their work, that is, whenever they had provided sufficient “sites” so that the absence of plaintiffs would not impede the safe progress of the mining.
If plaintiffs noticed that a mining machine was “off track,” that is, out of alignment with the “sites,” they would — and here the testimony of plaintiffs conflicted with that of witnesses for defendant — either report the need for correction to the mine superintendent or his representative or make the necessary correction themselves and later report the error and their correction of it. In order to insure greater accuracy, plaintiffs were required, with the aid of a second assistant, to take certain measurements every six months using a larger and, therefore, more accurate instrument than that which they used in setting the “sites.” In addition to these duties, plaintiffs were responsible for completing periodically certain maps and reports. Approximately daily they entered notations in a “field book.” From these entries the maps were prepared. While the testimony of witnesses for the defendant possibly conflicted with that of plaintiffs as to plaintiffs’ precise contribution to the preparation of the maps, plaintiffs indicated that their helpers transferred the information from the “field books” to the maps and that they themselves merely checked the accuracy of the helpers’ work. One of the maps was updated approximately weekly for purposes of safety; others, known as “progress maps,” which reflected the actual mining — as opposed to that proposed in the mining plan — were updated monthly. From their notations plaintiffs also computed and reported monthly both the total coal production and the tonnage of royalty coal.
While plaintiffs’ helpers were paid an hourly rate of compensation, plaintiffs received a straight monthly salary. Plaintiff Gilbert’s salary during the period in question ranged from $1,130 to $1,500 per month; plaintiff Wade’s salary was $15,800 for 1974 and amounted to $12,700 for that part of 1975 during which he was in defendant’s employ. Neither plaintiff had a college degree or was a registered surveyor or licensed engineer. Defendant’s manager of corporate engineering testified that at the time plaintiffs were hired as resident mining engineers, the *492“understanding” was that they would work whatever hours “it took to do the job. If they could do their job, if they could set the sites, do the work that was necessary to keep the mine in a safe position, in 30 hours, we had no objection to it, but if it took 48 hours they were also expected to do it. That was the understanding with the scale they were paid, or the salary they were paid.”
Like defendant, neither plaintiff had kept any written record of the hours he actually worked. Gilbert estimated that he had worked a weekly average of 50 hours; Wade estimated that he had worked a weekly average of 48 to 50 hours. Wade stated that he generally came up “on top” about 2:45 p.m. Gilbert stated that he sometimes came up “from the underground” about 2:30 p.m. and that he sometimes came up “a little earlier than that.” He indicated that he came up when all the underground work was done in order to work on “the books.” He said that the “field books” are not filled in underground because “[t]hey get too dirty and nasty. You get cold [sic] dust on ’em and they could get lost or destroyed underground.” Wade did not say where he was when he entered notations in the “field book.” Defendant’s manager of corporate engineering testified that, except for certain mapwork, “the actual bookwork was definitely to be done below. I heard testimony to other than that today.” The witness explained that “the sites and so forth ought to be put in the books below because of a chance of error from transposing [sic] from one to the other.” Both plaintiffs testified that 85 to 90 percent of all their work consisted of the surveying done underground.
Wade testified that during the period in question he had worked six days a week. In response to the inquiry of the trial court as to the hour at which his day at the mine began, the following colloquy took place:
“A. Well, my regular startin’ time, with different buddies, the majority of the time we’d start about, I’d say got at the mines around 6:30 to 7:00 o’clock and generally went below with the bosses at 7:15, but I have started at 4:00 and 5:00 in the morning, and I have been out at midnight, I’ve worked all shifts.
THE COURT: What was the occasion for your coming earlier or working later?
A. Well, it was just what needed to be done. If I couldn’t get it done or if there was something that needed to be done earlier, I figured that it was my duty to be there at that time, and sometimes just for my own convenience.”
Wade further testified that the mines sometimes were not worked on Saturday and that he had worked a “double shift a lot of times on Friday night,” usually when the mines were not being worked the next day but “sometimes” when they were, in order to have Saturday off “cause I liked it for my own convenience * ° He said on those occasions he would *493“come out, say, at 9:00 or 10:00 o’clock Friday night and work ’til 2:00, 3:00, 4:00 or 5:00 o’clock in the morning ’til whatever time it took me to get done.” Asked on re-cross why he wanted to be off, plaintiff answered,
“Well, I have — it’s well known in here I like to hunt, I like to fish; mostly hunt. I have worked my hours around, and Mr. Bailie [defendant’s manager of corporate engineering] knew it at the time, that I’d take off work hours that I could hunt a little bit.”
Defendant’s manager of corporate engineering, Cecil Bailie, testified in response to the question of whether he had checked to see how many hours plaintiffs worked daily or when they arrived or departed, that he had not, adding:
“And we weren’t supercritical of it, as long as he did his job. We knew many times that they left the mine early. Everyone has personal duties and so long as they kept their job in shape — kept their mine in shape — we had no objections.”
Gilbert did not testify as to the number of days he worked or the time at which his work day began or whether he too at any time worked a “double shift.” However, plaintiffs called William Scott, superintendent of the mine where Gilbert had worked at the time in question and retired at the time of trial. He testified that he had “no way of knowin’ ” how many hours this plaintiff had worked a week. Nevertheless, when asked, “Do you know whether he worked more than 5M or 6 days a week?”, the witness answered, “I’d say that they averaged 5M days a week.” This witness also testified that plaintiff Gilbert “[came] up” with him around 2:30 in the afternoon and “got down” with the bosses 30 or 45 minutes before starting time at 8 o’clock in the morning. Asked, “Did Jack [Gilbert] ask to get off early at any particular time?” he responded, “No, I’ve had Jack to come out of the night work so he could be off to go aflyin’ or something ° °
In their brief plaintiffs maintain that “it is uncontradicted that Jack Gilbert and Bill Wade worked more than forty hours per week.” The only testimony as to the number of hours each plaintiff worked was Gilbert’s estimate of 50 hours, Wade’s estimate of 48 to 50 hours and William Scott’s statement that they worked 5% days a week. They conclude that since defendant failed to keep any records of the time plaintiffs worked, defendant put on “no contrary evidence or impeaching evidence” and it was, therefore, error for the trial court to find that plaintiffs failed to prove the number of hours plaintiffs worked each week.
In a claim for unpaid overtime compensation “[i]t is incumbent upon the employee to show by a fair preponderance of the evidence not only that he worked in excess of the statutory work week, but the actual number of hours worked in excess thereof.” (Handler v. Thrasher (10th Cir. 1951), 191 F.2d 120, 122.) Not long after the passage of the Act, the *494Supreme Court addressed the problem of an employer’s failure to keep accurate or adequate time records and formulated the rule as to the consequences of such an omission:
“When the employer has kept proper and accurate records the employee may easily discharge his burden by securing the production of those records. But where the employer’s records are inaccurate or inadequate and the employee cannot offer convincing substitutes a more difficult problem arises. The solution, however, is not to penalize the employee by denying him any recovery on the ground that he is unable to prove the precise extent of uncompensated work. Such a result would place a premium on an employer’s failure to keep proper records in conformity with his statutory duty; it would allow the employer to keep the benefits of an employee’s labors without paying due compensation as contemplated by the Fair Labor Standards Act. In such a situation we hold that an employee has carried out his burden if he proves that he has in fact performed work for which he was improperly compensated and if he produces sufficient evidence to show the amount and extent of that work as a matter of just and reasonable inference. The burden then shifts to the employer to come forward with evidence of the precise amount of work performed or with evidence to negative the reasonableness of the inference to be drawn from the employee’s evidence. If the employer fails to produce such evidence, the court may then award damages to the employee, even though the result be only approximate.” (Emphasis added.) (Anderson v. Mt. Clemens Pottery Co. (1946), 328 U.S. 680, 687-88, 90 L. Ed. 1515, 1523, 66 S. Ct. 1187, 1192.)
An employer who fails to keep time records does so at his peril (Caserta v. Home Lines Agency, Inc. (2d Cir. 1959), 273 F.2d 943; George Lawley & Son Corp. v. South (1st Cir. 1944), 140 F.2d 439, cert. denied (1944), 322 U.S. 746, 88 L. Ed. 1578, 64 S. Ct. 1156), at the risk of having to disprove with little or no evidence that which was adduced on behalf of the employee (Skipper v. Superior Dairies, Inc. (5th Cir. 1975), 512 F.2d 409). Nevertheless, if an employee fails to produce evidence sufficient to show the amount and extent of that work, as expressed in Mt. Clemens Pottery Co. “as a matter of just and reasonable inference,” the' employer must disprove nothing to prevail whether he has or has not kept time records, and his failure to keep such records will be of no consequence.
Absent time records of the employer, courts have said that an employee may satisfy his burden of proof, enabling the court to determine approximate damages, if he provides information “such as the reasonable and creditable estimates of the employees themselves.” *495 (Brennan v. Parnham (W.D. Pa. 1973), 366 F. Supp. 1014, 1025; Brennan v. Carl Roessler, Inc. (D. Conn. 1973), 361 F. Supp. 229.) Mere estimates of hours of work performed, without more, are not, one may infer, “sufficient evidence to show the amount and extent of that work as a matter of just and reasonable inference,” as required by Mt. Clemens Pottery Co. We note that in both Parnham and Roessler the results of an investigation conducted by a compliance officer of the Labor Department supplemented the employees’ estimates. Furthermore, in the case at bar, unlike the courts in those two cases, the trial court expressly stated that there was not enough evidence in the record to determine whether plaintiffs’ estimates were reasonable and creditable.
Although there was testimony in the instant case with regard to the time plaintiffs usually came up from their underground work, it can not be determined from the evidence when plaintiffs’ work day actually began. While Wade indicated that he arrived at the mine as early as 6:30 a.m., William Scott testified that starting time was 8 a.m., 1M hours later. Although Scott stated that Gilbert went down into the mine at 7:15 a.m. and Wade indicated that he too went down at that time, there is no evidence as to how long it took plaintiffs to reach the location of their work or as to what tasks, if any, plaintiffs performed primarily for the benefit of defendant during this 45-minute period, assuming that the trip down into the mine did not consume the entire 45 minutes. Since it is not known whether plaintiffs had any idle time between 7:15 and 8 o’clock, there is no need to determine whether any idleness was occasioned by a requirement of defendant, as it might have been, for example, had defendant required plaintiffs to descend into the mine by a 10-minute trip starting at 7:15 and to wait until 8 o’clock before beginning their underground work. Nor was there any evidence as to whether either plaintiff took a lunch break during any of the time he spent underground. Once again, since there is no evidence with regard to the matter at all, it is unnecessary to know whether defendant’s other employees, for example, were permitted to consider their lunch period as working time. Since there is no means by which to determine how many hours plaintiffs actually worked underground, there can be no way of knowing how many hours constitute 85 to 90 percent of the total number of hours worked. Nor, as a result of that omission and Wade’s failure to provide a time when he finished working on “the books,” can it be known how many hours constitute the 10 to 15 percent of working time spent in that way.
While Gilbert did not testify that he ever worked a “double shift” on Friday as Wade did, Wade testified that he would “come out” at 1 or 10 p.m. and leave as early as 2, though as late as 5 a.m. If he started his work as late as 10 and left as early as 2, he worked only a four-hour day. By the same token, if he started at 9 p.m. and left at 5 a.m., he worked an eight-*496hour day. There was, of course, no indication whether he took a break to eat during the night shift. If one were to assume, for example, that Wade took a half-hour lunch break when he worked during the day, which started at 8 a.m. and lasted until 2:30 p.m., Wade would have worked but a six-hour day underground. If one were to add as much as a six-hour shift on Friday to five six-hour days, that would be but a 36-hour work week underground. If he worked four hours on “the books,” which is a little more than 10 percent of 36 hours, he would still not have been eligible for any overtime compensation. Since he claimed to have worked an average of 48 to 50 hours a week for something over 18 months, a reasonable trier of fact could have found that plaintiff Wade had not proved the amount of overtime to which he was entitled. Since Gilbert testified neither to the number of hours he worked each day — and gave no means of inferring that number — nor to the number of days he worked each week, the trier of fact could well have found that he failed similarly in his proof. That William Scott stated that Gilbert “averaged 514 days a week” is less persuasive than it might have been had it not followed almost immediately the statement of the witness that he had “no way of knowin’ ” how many hours a week Gilbert had worked during the period of time in question.
Therefore, for the foregoing reasons the judgment of the circuit court of Franklin County is affirmed.
Affirmed.
KARNS and KASSERMAN, JJ., concur.