Arkansas Power & Light Company, appellant herein, brings this appeal from a judgment entered in the Hot Spring County Circuit Court in favor of numerous landowners who successfully contended that their lands had been damaged by the negligence of appellant company in its operation of Remmel and Carpenter Dams. The judgments totaled $18,541.07. Three points are urged for reversal, including alleged erroneous instructions, but under the view that we take, it is only necessary that we discuss the first point, vis., that the trial court should have directed a verdict in favor of the company.
On the morning of June 16, 1963, between the hours of 2:00 and 3:00 A.M., it began to rain in the area of Carpenter and Remmel Dams and continued until approximately 12:00 noon on the same date. Total rainfall at Remmel Dam (which formed Lake Catherine) during this period was 6.14 inches, and the total rainfall at Carpenter Dam (forming Lake Hamilton) during this same period was 4.56 inches. The maximum elevation at which water may be contained in Lake Catherine is 305 feet, and the maximum in Lake Hamilton is 400 feet. Both dams contain flood gates for the purpose of channelizing the water.
Appellees, in their complaint, alleged that appellant on July 16, 1963, carelessly and negligently raised flood gates under its control, and permitted excessive water to flow into the Ouachita River below Remmel Dam, causing such water under its control to overflow lands on which the appellees had crops, destroying and damaging such crops. It was specifically asserted that appellant failed to obtain weather reports from the United States Bureau showing rainfall predictions, and that *461such failure amounted to negligence. Their main contention was set out in their answer to the company’s Interrogatory No. 4, wherein the question was propounded relative to how the Power and Light Company held back water and prevented it from flowing naturally down the Ouachita River. The answer to this interrogatory states:
“It had water stored held back by its dams prior to any Act of God causing an unusual inflow. After discovering and recording the rise coming on its lakes, including the rate of rise, it delayed opening of gates to let the water flow down the river naturally and then after seeing that it was getting more water than could be handily handled, it opened its gates and turned water down the Ouachita River in a larger volume than would have gone down the river naturally. After releasing sufficient volumes of water to cause a flood on the lands below the dam. it continued to release larger volumes of water than were naturally flowing into its lakes, lowering the water level of its lakes while so doing and thus held flood waters on crops of the plaintiffs a longer period of time than the water would have remained thereon by natural flow.”
The lands of the appellees are located in the Ouachita River bottom area below Rockport, near Malvern, and are downstream from both Remmel Dam, and Carpenter Dam, which is about 11 miles farther away. Elevation maps reveal that there is a 95-foot drop from Carpenter to Remmel Dam and a 47-foot drop from Remmel to Rockport. There is a further drop from Rockport Bridge to Morrison Island where most of the appellees were farming) of 39 feet, making a total drop in elevation from Carpenter Dam to Morrison Island of 181 feet. It is thus apparent that the lands of appellees are very low.
One of the principal witnesses for appellees was Lewis Cash, who lives in the Midway community, and *462has farmed in the Ouachita River bottoms for many years. Cash has observed various floods in the vicinity throughout the years, and the witness testified that the flood in question was different from the others in that the water remained on the lands longer than usual. “Normally when it crests, it peaks, and within 3 or 4 hours it goes to falling pretty fast. This time it hung on when the main part of the water left there.” Cash said that his land remained under water for 40 or 45 hours. The witness stated that when flood waters run off quickly without the property being subjected to a hot sun, there is but little effect on the crops; however, if the water stays on for a long period of time under hot sunshine, the crops will sour. Cash said that the 17th of June was a “real hot day,” and that the water was still over the crops at that time. This appellee testified that 10 or 12 hours of hot sun would ruin the crops, but that water staying on overnight probably would have no adverse effect. “I don’t think it would just staying-on it overnight, a short period of time, if it was cool.” He “couldn’t say definitely” what time period alone (for the water to remain on the crop) would kill the corn.
Doyle Cook, a farmer in the area, received the report of a flood about 8:00 A.M., and he testified that he forthwith checked various creeks and streams, some running into the Ouachita River below the dam at Lake Catherine, and some of the streams running into the lake above the dam. According to the witness, most of the creeks and streams were “normal,” and others indicated that they might have been a little out of their banks (because of debris that was observed), but the witness found no creek or stream out of its banks. This included Tigre Creek and G-ulpha Creek, and this evidence is rather strange in light of the testimony of other witnesses which will subsequently be quoted. As to Tigre Creek, Cook stated that it had been up some, but was down when he crossed it around 10:30 or 11:00 A.M.; there was no debris in the road or highway at Grulpha *463Creek. Cook even stated that there were places where it appeared there had been no rain at all (Cooper Creek, which empties into Lake Catherine), and there were other creeks where the rain did not “amount to anything.” Some of this testimony is confusing, but the purport of his evidence was that the water, about which the appellees complained, came from Remmel Dam.
E. C. Stewart, presently retired, but at the time of the flood, Systems Operator for the Middle South System, and whose office was in Pine Bluff, was called by both appellees and appellant. Mr. Stewart had control over opening the gates, including the determination of which gates would be opened, at both Remmel and Carpenter Dams. The witness said that the drainage area between Blakely Dam and Carpenter Dam was 300 square miles; from Carpenter to Remmel was 120 square miles; in other words, all the water in this area flowed into Lakes Hamilton or Catherine. He testified that on a normal day, his office would be in touch with personnel at the dams not less than once an hour, but that when something unusual happened, communications were increased. Flood records for July 16 were introduced, and these reflected the following facts:
When the water level in Catherine exceeded the 305-foot elevation, employees of the company began opening the gates, the first opening being 4 feet, and occurring at 6:08 A.M. At 8:47 A.M., the same action occurred at Lake Hamilton. Subsequently, the gates at Remmel were opened to 10 feet.
The evidence offered was voluminous, but a summary reflects that at Remmel Reservoir for the 24-hour period from 12:01 A.M. to 12:00 midnight on July 16, 48,934 acre feet came into the reservoir, and 45,085 acre feet were discharged. On July 17, 11,325 acre feet flowed into Remmel, and 14,792 acre feet were discharged. In other words, during this 48-hour period of July 16 and 17, the total inflow was 60,259 acre feet, *464and the total discharge was 59,877 acre feet; i.e., approximately 400 acre feet more came into Lake Catherine than went ont. On July 16, 28,116 acre feet flowed into Lake Hamilton and 18,490 acre feet flowed out. Prom midnight until noon on the 16th, more water flowed into Hamilton than was discharged at the dam, but from noon on more water was discharged than came into the lake. There were no figures on the intake and discharge at Hamilton on the 17th. A flood record sheet for Carpenter Dam on the 17th was not kept because it was considered that the flood was over.
Appellees contend that the power company, although it had the capacity to do so, did not maintain the smoothest possible flow of water; that though appellant at Bemmel Dam had the gate capacity to let out all water coming in, as it came in, the company failed to do this, but instead, held the water, discharging it hours later. It is asserted that appellant should know that:
“* * * unnecessarily discharging water into the main channel of the Ouachita Biver at a time flood water was out over the crops would hold the crops under water a longer period of time and thereby destroy them, when if the water had come on through naturally, even if it peaked at a higher point, would have cleared from the crops much sooner than it did in this case.”
Unquestionably, the rainfall at the sites of the dams, and the general surrounding area, was unusually heavy, conditions being described by some witnesses as being the worst that they had seen.
G-len Teeter, a vocational teacher and resident of Magnet Cove, testified that when he awakened around 6:00 A.M. on the morning of July 16, it had rained to the point that Stoney Creek, which flows through his field and empties into Ouachita Biver about 4 miles below Bemmel Dam, was up enough that the creek ran knee *465deep through a neighbor’s house which was just across a road adjoining the creek; he also determined the amount of rain by checking a 5-gallon bucket that he had emptied the night before and left in the back yard. Teeter stated that the bucket had straight sides and was 12 inches deep, and that on the morning of the 16th it was full. He said that it did not rain prior to his going to bed on the night of the 15th which was 9:00 or 10:00 P.M. The weather record at Malvern shows rainfall of 5.20 inches on the 16th.
Barney Roark testified that there had been a big flood on G-ulpha Creek (which empties into the upper end of Lake Catherine) on the morning of the 16th, and that the water flowed over the bridge on Highway 270, and washed away the Rock Island Railroad tracks for % of a mile; big gum trees were “snapping” as though they were toothpicks. Roark owns a building approximately 40 yards from the creek, with a 7-foot ramp built around it, and the water reached 24% inches in this building. The witness also lost other property in the flood, including a trailer parked on his property: “It was a big show trailer and well equipped, and I had a new, brand new Evinrude motor in there, and I haven’t found one part of it yet.” This occurred around 7:00 A.M.
Yal Hall, Superintendent of the Municipal Water System at Hot Springs, testified that he was called to the plant about 8:00 A.M. on the 16th because there were indications that the dam used for the Hot Springs water supply (Lake Dillon) might wash out. The water was running over the dam about 5 feet deep:
“Well, on the way out I had a little trouble getting through town. Town was flooded down Central Avenue. As soon as I got out to the plant, the -engineer wanted me to come out there. I don’t know what I could have done if it had washed out. But anyway, water was in the end of the pump-*466house. We never experienced anything like it on that watershed, not since we have been there. When I got out there, water was everywhere and I asked how much rain we had since midnight. We always read the gauge at midnight. We went out to the rain gauge and checked the rain gauge and had 9.98 rain since the meter was read at midnight. It was about 9:30 or ten o’clock a.m. when I read the meter. ’ ’
The witness stated that the next day he went down to the basin known as Bull Bayou, which empties into Lake Hamilton, to see what damage had been done.
* * But in going down to the basin, it looked like a dozer had gone in down there. Trees were washed up. I never seen trees with bark peeled on them before, or since. It had washed up trees, washed bark off. It was really a fierce deal down in that basin. * * *
“On Grulpha Creek we had ten inch cast iron pipelines across Grulpha Creek and it washed three lengths out of it. One length we found about six months later, on down stream where it lodged on a little dam on Grulpha Creek. Part we never did find. But, there was three lengths that washed out. ’ ’
These lengths were of 10-inch inside diameter, 18 feet long, and weighed 1150 pounds each. Mr. Hall testified that a daily log of rainfall was kept at the pump station; that the record reflected no rain at all on July 15, but on July 16 there was recorded 9.98 inches.
Ernest Echols testified that in 1963 he lived above Remmel Dam near Wilson Mineral Springs Creek (also known as Potash Sulphur Creek), which empties into Lake Catherine. Echols testified that he had a boat *467landing at the point where this creek runs into the lake, and that this boat dock was severely damaged by the flood:
“Well, when I got home from work right after seven o’clock that morning, during the flood, when I drove into my landing, I seen the boats was all washed away and the boat dock, big dock, had washed out and my gasoline pump washed away.”
This occurred about 7:30 A.M. on the 16th, and Mr. Echols also testified that on his way home he crossed Tigre Creek, which also empties into Lake Catherine, and that this creek was the highest that he had ever seen it — “way out of banks.”
“The Pierson house sets on the right just before you get to the bridge. Water was running in the windows of the house, and a car setting out in the yard, water was running in the glass on the door of the car. That’s how high it was.”
As to the force of the water, the witness said:
“Coming down, well, it was real high and a lot of force and the water coming in that G-ulpha Creek, coming in at my landing, comes under the railroad trussle, this big bridge, they have a trussle, it was round about twenty foot of water was coming through that and hitting the lake water in such force, it was like a water spout or a nozzle, bouncing as high as this ceiling, washing my boats and bouncing through there. Water was bounding fifteen feet or higher from the creek.”
Echols said the water was up all day; that he had lived in the area for 22 years, and had never seen Gulpha or Potash Creeks up to the extent that existed on this occasion.
*468Don McGrew, who operated a store and court near Glazypeau Creek (which empties into Lake Hamilton) was called to testify, and counsel for appellees stated:
“We will agree there came the biggest rain he had ever seen since he had been living up there and the biggest flood on Glazypeau since he had been there and it washed his cabins off and he hasn’t seen them since and also the water went down into Lake Hamilton.”
When asked where he was during the flood, McGrew replied:
“Part of the time on top of them [cabins] trying to help a man get out.
“It was washing a big long trailer away on the east end of the cabin and a fellow was there that couldn’t swim out and I was helping him. We decided we better get him on top.”
The witness stated that the water rose to approximately 6 feet inside the cabins which are about 75 yards from Glazypeau Creek. This occurred about 8:00 A.M., and McGrew said that it rose about 10 feet from daylight until that hour.
County Judge Lon Warneke testified that the heaviest rains in Garland County were from the Mountain Pine area, east, and between the town and the mountain range just north of Hot Springs, and that all this water emptied into either Lake Hamilton or Lake Catherine. It was the heaviest rain, and more damage was suffered, than at any time since Warneke had been County Judge.
It appears that it would be extremely difficult to find in the record sufficient evidence to make a jury question on the issue of negligence. In Ark Power and *469 Light Company v. Beauchamp, 184 Ark. 698, 43 S.W. 2d 234, also a case involving Carpenter and Remmel Dams, the court commented that employees at Remmel Dam had means by which they could communicate with the weather bureau, and thus obtain information concerning rainfall and the volume of water likely to come down Ouachita River, but that this precaution was not taken. In the present case, there was an effort to obtain information from the United States Weather Bureau at Little Rock, but this agency had no information relative to rainfall in the Hot Springs area (which it had not predicted) and, in fact, the weather bureau asked the company for information. Appellees state that there is no evidence that company employees endeavored to utilize the services of television, radio or otherwise obtain other weather information — but there is no indication of what would have been shown or disclosed by such information. Apparently, the rainstorm was of a very sudden nature, taking .everyone by surprise. The company was confronted with an emergency. Who can say that, if the water from the creeks and streams, described in the testimony heretofore set out, had been permitted to go immediately through the dams, that an even greater flood might not have resulted?
In Beauchamp, the court said that the circumstances warranted the inference that the water came from Lake Catherine, and that the flood gates had been opened negligently, thus precipitating within a few hours the water which had been flowing naturally, and more slowly, downstream. We held that the jury was justified in concluding that the company had opened the flood gates more than necessary, and to such an extent that the flood resulted. Here, actually, the contention of appellees is in reverse, for they state that, if the company had permitted the water, occasioned by the terrific rainfall, to flow unimpeded, the water would have moved much faster, and though perhaps reaching a higher crest, would still have flowed off more quickly. Mr. Stewart, who had long years of experience in operating *470dams, and Colonel Charles Maynard, who has spent 24 years with the Army Engineers (the latter conceded by appellees to be an expert on flood control and the movement of water), testified emphatically that the manner in which the gates were operated under the circumstances of July 16, was entirely proper. Colonel Maynard stated that the water removal from the dams was consistent with the best principles. It was his view that the lands belonging to appellees received less damage than would have been done had the two dams not been in existence.
There really is no necessity in discussing at length the question of negligence, for even though it should be established that appellant’s employees were negligent, there is another factor which is of equal or greater importance, vis., whether the negligence of the Power and Light employees was the proximate cause of the damages suffered by appellees.
As to this phase of the litigation, appellees’ claim for damages is wholly predicated on the theory that the company, by holding the flood waters for several hours and then letting the water out in the amount heretofore set out, caused water to remain on the lands for a longer period of time, permitting the crops to be exposed to long hours of hot sun, and thus causing them to sour and ruin.
Lewis Cash testified that his corn was under water for 40 to 45 hours (entirely over the stalks of a lot of the crop), and that 10 or 12 hours of hot sun would ruin the crop. Cash testified that a few years back, the water came over his corn one evening, and went off the next morning, but he made a “fair” crop.
Bill Cook testified that he could not say exactly how long it would take for water to kill a crop, because of the fact that temperature would have a lot to do with it. The witness said that it was hot on July 16, and *471that he would estimate that it would take about 45 hours to ruin the crop. This estimate was apparently based on the fact that the water was on his crop for that period of time, but he had no idea whether water, standing for a lesser period would likewise have ruined the crop. This answer was given relative to his corn crop, but he said that the same answer would apply to soy beans and hay. The basis of the opinion of the witness was the fact that he had been flooded before, but the water had not remained on his land for that long a period of time.
Doyle Cook stated that the water was off the bulk of the crops by the night of the 17th.
C. C. Bozeman testified that his corn, which was in the tasseling stage, was entirely covered by the water, but he was unable to say how long water would have to stay on corn before it would be damaged.
Glen Teeter testified that corn begins to be damaged after water stays on it for 5 or 6 hours, and that in 10 or 12 hours the crop would probably be destroyed. He was of the opinion that a soy bean crop would be destroyed within 10 hours.
It will be observed that the witnesses pretty well agreed that the flood waters remained on the crops from 40 to 45 hours. However, the testimony reflects that if the lands were under water for a much lesser period (10 to 12 hours under hot sun, according to Cash; 10 or 12 hours under water, according to Teeter), the crops would be totally destroyed anyway. In other words, even though appellant’s operation of its dams caused the water to remain on the crops for a considerably longer period, the company would not be liable if the water would have remained on the lands long enough to ruin the crops, had it been permitted to flow unimpeded through the dams. Thus, the 40 to 45 hour period would appear to be immaterial, for the maximum damage had been reached many hours previously. The *472question therefore is, “Did the manner by which appellant company released water from the dams cause the crops to remain under water for the first 15 or 20 hours after being flooded?” Here, we come upon a blank wall — for there is no evidence that this was true. Prior experiences of the witnesses have but little evidentiary value for the reason that it seems established that previous floods did not compare with this flood in magnitude.
Stewart testified that this was the heaviest and most sudden downpour of rain that had been experienced since the dams had been constructed, and the rise was much sharper than prior rises.
Appellant presented a graph prepared by Stewart1, the purpose of which was to show that the flood would have been greater, i.e., much more water would have gone below Remmel Dam if the Ouachita River had flowed uninterrupted. According to the witness, the presence of the dams prevented the flood from reaching the peak that would have occurred had the dams not been present.
From the record:
“Q. In regard to the rate of flow that yon have talked about on this chart, will you state whether or not — Let me ask you this: What was the measurement in acre feet per hour that would have gone down had there been no dam at that particular point as shown on your chart, Exhibit No. 26?
A. The highest point is 8,653 acre feet per hour.
Q. What is the feet, the discharge as it actually occurred?
*473A. That occurred at 10 A.M. and was 4,655.”
Further:
Q. What was the time interval from the time, beginning of the flood in the Ouachita at that point without any dam, the red line, ’til it got below flood stage at that point?
A. It was eighteen hours.
Q. What was the time intermal from the time, with the two dams, it got in the flood stage to the time it got out of flood stage?
A. * * * Eighteen on the red line without the dams and twenty-one on the blue with the dams.”
At any rate, for appellees to establish their contention that appellant was responsible for the loss of the crops, it must be shown that the water would not have remained on the lands long enough to occasion the damage except for the actions of appellant. Such a conclusion can only be reached through speculation, for there is absolutely no evidence that this happened — nor is there any evidence that would tend to support this as a reasonable conclusion. In Kapp v. Sullivan Chevrolet Company, 234 Ark. 395, 353 S.W. 2d 5, we said:
“* * * Appellants’ entire case rests upon conjecture and speculation. Several possible causes of the break are argued, but in truth, there are only possibilities, and do not reach the status of probabilities.”
In that same case, quoting from Henry H. Cross Company v. Simmons, 96 F. 2d 482 (a decision under Arkansas law), we said:
*474“ ‘To submit to a jury a choice of possibilities is but to permit the jury to conjecture or guess, and where the evidence presents no more than such choice it is not substantial, and where proven facts give equal support to each of two inconsistent inferences, neither of them can be said to be established by substantial evidence and judgment must go against the party upon whom rests the burden of sustaining one of the inferences as against the other.’ ”
Here, it is not sufficient to show that the crop damage could have been caused by the manner in which the water was released from the dams. Proximate causation must go far beyond guesswork. Who can say that the damage would not have occurred if the water had been permitted to flow unhampered and unhindered? Unquestionably, more water would have descended upon appellees, and logic dictates that the larger the body of water, the longer it would take it to run off. Who can say — or who did say — that, except for the manner of operation of appellant’s dams, the water would only have stayed a few hours? To this most pertinent question there is a total absence of substantial evidence. That being true, appellees must fail.
Reversed.
Jones and Byrd, JJ., dissent.
I am of the opinion that the appellees made out a jury question in this case even though the negligence and causal relationship border on the speculative.
It appears that there is little question that appellees suffered damage to their crops because of high water. The alleged damage was not based on the volume of water covering the crops; the alleged damage was based on the length of time the water remained on the crops.
*475Piad there been no dams maintained by the appellant on the Ouachita River, the appellees would have been confronted with a situation commonly referred to on the Ouachita River as a “head rise.” In the normal flow under such conditions as is evidenced by this case, the water would have flowed down the Ouachita River basin in one tremendous volume, receding as fast as it rose, carrying away obstructions and leaving debris in its wake, as occurred on Grulpha Creek and in the Bull Bayou basin under the testimony in this case.
So it appears to me, that the appellant was confronted with two alternatives; it could release the water from its reservoir in the same volume and over the same period in which it fell, thereby permitting a higher, faster and possibly more devastating flood below Remmel Dam on the Ouachita River; or it could release less water than fell in the first twenty-four hours and more water than fell in the second twenty-four hours, thus regulating and controlling the height of the flood below the dam but maintaining it over a longer period of time.
I recognize the speculative nature of what the damage might have been had the appellant released the flood waters in the same volume and over the same period they were impounded, but it is my opinion that all of this was presented to the jury under proper instructions so I would affirm the judgment rendered on the verdict.
I am authorized to state that Byrd, J., joins in this dissent.