Opinion by

The argument of the appellant directs our attention first to the complaint variously set forth in the sixth, seventh, eighth and ninth assignments that the case was submitted to the jury without evidence on the part of the plaintiff that the defendant’s bridge was negligently constructed or that the injury complained of was attributable to any fault therein which ought to have been avoided. The right of the defendant to maintain a bridge over the stream is not questioned, nor can it be disputed that there is a presumption of care in favor of a railroad company which constructs a bridge in the exercise of its franchise as a public corporation. It does not follow, therefore, that the plaintiff is entitled to recover merely from the fact that the stream was obstructed by the formation of a dam *239and the consequent overflow of his land. No testimony of experts was offered by the plaintiff bearing on the alleged neglect to build an adequate bridge; but there was evidence on which the court was justified in submitting to the jury the question whether the defendant had performed its duty in this respect. A former bridge in the same location had been burned, and the structure complained of was erected for temporary use. The width of the stream was about 220 feet and there were located in that space nineteen or twenty piers composed of piles upon which stringers for the support of the ties were placed. These piers, according to the evidence presented by the plaintiff, were set at an angle of about thirty degrees with the line of the current at usual high water and the spaces between the piers were estimated by different witnesses to be from eight to fifteen feet. The spans of the permanent bridge built to replace that which was burned were sixty feet in length. Common observation would suggest that a structure of this character in a stream like the Cowanesque river might be an obstruction to the flow of water and the accompanying drift in case of floods, and it would not seem to be necessary that experts be called to satisfy a jury of that fact. The action of streams at high water and the requirements of ventage capacity at bridges are so generally known that defects of construction in some instances may be obviohs from a description of the bridge and the stream, and in this case we think the jury might properly determine from the evidence whether the conditions were so plainly dangerous as to admit of the inference of negligence. The defendant was bound to anticipate and provide for the magnitude of ordinary floods with their usual ice and drift. It was not bound to foresee what was extraordinary; provision for the contingencies of ordinary floods would be a fulfillment of its obligation. The size of the stream, the height of its banks and the extent of the water shed were conditions of which the defendant was bound to take notice. The situation called for the exercise of the requisite skill to adopt a plan necessary to avoid obstructions to the flow of the stream in ordinary conditions and the untoward consequences of the formation of a dam by the accumulation of ice or drift in ordinary freshets. *240Whether the floods by which the plaintiff was injured were extraordinary or usual was a matter of dispute at the trial, but there was abundant evidence from which a jury might find that the flow of water was not greater than was frequent in that stream, and at the occurrence of each of the floods in January and July of 1902, the stream was so obstructed at the piers, in the first instance by ice and in July by hay and other drift, that a large volume of water flowed out of the channel and onto the plaintiff’s land. The court submitted to the jury to find whether the floods were extraordinary and whether the bridge was adequate to permit the flow of the water in the channel at ordinary stages of flood, añd the verdict of the jury was against the defendant on both of these questions. It must be taken as established, therefore, that sufficient provision was not made for the stream when the structure was erected. In the view which we take of the case there was' evidence supporting this finding. The case of Berninger v. Sunbury, etc., R. R. Co., 203 Pa. 516, cited by the appellant, grew out of an entirely different state of facts from that now before us. There the bridge was a permanent structure and was shown to be sufficient for all ordinary conditions of the stream; it had three piers in a length of 182 feet. The stream was about fifty feet wide and had a rapid fall. The plaintiff’s testimony showed that the vents in the bridge had ample capacity for any ordinary flood and any ordinary ice formation on the surface, and that while there had been floods in the creek there had been none since the bridge was built as bad as the one which caused the injury. In the present case no attempt was made to show that the bridge was properly constructed considering the character of the stream, and all of the plaintiff’s evidence goes to show that it amounted to a serious obstruction.

The second, third and fourth assignments complain of the answers of the court to the defendant’s first, second and sixth points which are predicated of evidence that part of the injury to the plaintiff was caused by water which flowed onto his land from the stream at a point below the bridge, and requested the instruction of the court that the plaintiff could.not recover for damage thus caused. The court refused the points as appli*241cable to the case because of evidence introduced that the presence of the ice and drift at the bridge forced the water toward the east end of the bridge and this water poured through two or three of the spaces at that end in such volume that it struck the east bank below the end of the bridge and washed a channel in the bank through which it flowed onto.the Burlingame farm and thence onto the land of the plaintiff. The jury was instructed, however, that the plaintiff could not recover for any injury caused by water which flowed from the channel below the bridge unless they should find that such overflow was caused by the change of the course of the stream owing to the presence of the ice and drift above the bridge. This instruction was not erroneous as there was evidence tending to show that the effect of the obstruction at the bridge was to change the course of the current and give it a direction below the bridge as indicated in the answer of the court to the plaintiff’s first point.

Objection is made to a portion of the charge of the court as set forth in the tenth assignment with reference to evidence on the measure of damages. The plaintiff proceeded to show the amount of his damage on the assumption that there was a permanent injury to the land, and the witnesses were asked without objection by the defense what the amount of that damage was. Testimony had been previously introduced that the plaintiff’s two fields on the north and south sides of the railroad had been badly gouged and washed by the floods; that the top soil was carried away and long excavations from one to five feet deep made in the earth. One of the fields contained about three acres and the other about six, and the descriptions by the witnesses of the injury done showed a serious permanent injury to the land. What the plaintiff proved was the actual damage to the land. This was the method adopted in Weaver v. Berwind-White Coal Company, 216 Pa. 195, and is supported by that and other cases. Where the injury is irreparable depreciation in value of the property is the measure of damages. If the injury be remediable but the cost of repair would be greater than the diminution in the market value, the latter is the correct measure. Compensation for the loss is that to which the plaintiff is entitled. The witnesses for the plaintiff were not *242questioned expressly as to the value before and after the injury, but the course of inquiry shows that their estimates of the damage were made on that theory. They were familiar with the situation and the value of land in the vicinity and based their estimates on that knowledge. Whether, therefore, the rule applied in McGettigan v. Potts, 149 Pa. 155, and Noonan v. Pardee, 200 Pa. 474, that the actual damage was the measure of the plaintiff’s right of recovery or that the difference in market value before and after the injury determined the amount to be recovered as applied in Fulmer v. Williams, 122 Pa. 191; Williams v. Fulmer, 151 Pa. 405; Thompson v. Traction Co., 181 Pa. 131, the plaintiff’s evidence enabled the-jury to find the amount of the actual loss. The defendant introduced evidence tending to show that the land might be restored by cultivation to a condition equally as good as that which had existed before the floods; that its fertility could be renewed and that the plaintiff’s damage was the amount which it would cost to produce this restoration. The comment of the court as set forth in the tenth assignment related to this evidence and amounted to nothing more than an observation that the defendant’s witnesses made the basis of their estimate of damages the cost of putting the land in the condition it was in before the floods. This did the defendant no harm. Indeed, the defendant’s witnesses did not testify that the land could be put in its former condition. What it proposed to show and did show by some witnesses was that it would be practicable to restore the surface of the land to a condition equally as good for farming as existed before the flood and that the fertility of the washed land could be restored to a condition equally good with that of the remaining surface soil. This was a very different proposition from that of showing what it would cost to place the land in its former condition. It was not seriously contended on the trial that this could be done. The injury might be amended and it was the cost of such amendment which the defendant showed. If the plaintiff could not be placed in his former condition and the defendant caused his injury he is entitled to damages equal to the actual depreciation in the value of his land.

We do not find reversible error in the admission of the evi*243dence of the witness, Wagner. He had lived within two and one-half miles of the land for forty years; he drove by it frequently on two roads and saw it right after the flood. He had some knowledge therefore, as to the extent of the plaintiffs injury. It will be observed, moreover, that his estimate of damage was lower than that of the other witnesses for the plaintiff and his lowest estimate much less than the verdict. The offers of proof by the witnesses Matson and Kiley as set forth in the thirteenth, fourteenth, fifteenth, sixteenth and seventeenth assignments were properly rejected. The proposition was to show by these witnesses that the surface of the land washed could be made so it could be worked as easily as any portion of the farm and be brought to as high a state of productiveness as the corn and tobacco fields. The witnesses had not seen the land before the floods and did not have any such knowledge of the condition of the soil as to qualify them to speak in regard to its fertility. Moreover, it was not proposed to put the land in the condition in which it was before the floods, and the offers were not competent for any purpose in the issue tried.

Charles M. Wood was called as a witness for the defendant and testified that he was acquainted with farm values in the vicinity and that the market value of the plaintiff’s land was $100 an acre. On cross-examination he stated that he knew of sales of land in the neighborhood. He was then asked if he did not know of a sale of tobacco land just below Knoxville in that neighborhood at $4,000 for seven acres. He admitted that the plaintiff’s farm was a good tobacco farm before the floods. This inquiry was competent for the purpose of testing the knowledge of the witness and ascertaining the basis on which he fixed the value of the plaintiff’s land. We think the appellant has not presented any substantial complaint as disclosed by the record of the trial.

The judgment is therefore affirmed.