Appeal from a jury verdict in a condemnation suit. Remanded for a new trial as to damages only. Costs to defendants.

The only point of moment on appeal is whether the trial court erred in precluding one Nielsen, appraiser called by defendants, from testifying as to the market value of the subject property. We believe and hold that under the circumstances of this case there was such error requiring a new trial as to damages.

The record is very voluminous, but for the purpose of this decision we feel that the pertinent parts thereof can be abstracted fairly and accurately as follows :

Nielsen was an expert real estate appraiser as shown by question and answer. No one questioned his qualifications as such. Pie testified he was acquainted with the subject property — two tracts containing about 156,000 square feet, situate across the street from a high school to the west, and across the street from a church to the north. Everyone agreed that damages were to be determined as of June 28, 1960. On direct examination Nielsen said the highest and best use of the property was for a multiple apartment complex. He testified that he 1) knew of two sales of comparable property in 1962, and one in 1961, the exact dates of which are not evident; 2) that he was aware of some real property sales in I960, but did not consider them comparable, learning of none during that year that were comparable to the subject property; and 3) that there had been no appreciation in real estate values during the two years mentioned.

We think that obviously, logically and legitimately, Nielsen should have been allowed to give his opinion as to the value of the property on June 28, 1960, after the testimony given. He was precluded from doing so by the trial court. Some of this testimony was elicited by a so-called “voir dire” examination by plaintiff’s counsel, engendered by his request for such an abortive procedure immediately after Nielsen was asked, on direct, to give his opinion as to value. Defendant’s counsel did not object to the ensuing searching inquiry under the so-called “voir dire,” which really was premature cross-examination; and in so failing, allowed counsel for plaintiff to really emphasize the propriety of admitting Nielsen’s testimony as to value. Plaintiff’s counsel cannot complain that defendant’s counsel waived the error of “voir dire,” since the former waived any objection to what he himself brought out on the misnomered “voir dire” cross-examination.

The trial court opined that Nielsen’s opinion as to value would be speculative. With the testimony Nielsen gave, we can see no reason why such opinion would be any more speculative than that of any other expert, testifying as did Nielsen, where factors of time, situs, terrain, other sales *343or absence thereof, etc. all present at least an aspect of individual and/or collective speculation.

If Nielsen was out in left field in his appraisal of the corpus of the property and its value, it was for the condemnor — he who was seeking to take the property- — to prove on cross-examination, at the proper time, that Nielsen did not know what he was talking about — otherwise plaintiff should have struck him out when he first was being examined as to his expertise.

The trial court may have had a labor pain or two induced by an unnecessary protraction of a very simple case that took over 600 pages of transcript to establish its simplicity. We are sure that the trial court really did not mean it when it said that “Having taken only 1962 prices, and without having considered anything in 1960, the court is of the view that Nielsen considers the 1962 prices more favorable to the landowner here who called him to testify.” The implication of dishonesty on the part of the witness is distasteful enough, under the evidence here; but it is more unforgivable for the trial court, in the middle of direct examination, punctuated by an erroneously permitted, but premature cross-examination, under the aura of the glittering magic words “voir dire,” to evaluate the credibility of the witness, which, at that juncture, was none of its business, but that of the jury.

After Nielsen’s opinion was rejected, Plaintiff called witnesses who testified throughout dozens of pages that Nielsen’s testimony was no good. Over half the plaintiff’s brief on appeal was devoted to the same kind of impeachment of what Nielsen had said, which completely was disusive and emasculated because he was prevented from making an answer to the only question that would make any sense or reflect the basis for his conclusion. Hence, the error of refusal to permit Nielsen to state his opinion as to value was fait ac-compli as of the time of such rejection, and if it were error at that time, ten thousand words or ten thousand throats thereafter could not purge the error.

Therefore, the rejection of the opinion evidence as of the time it was proffered is controlling in the opinion of this court, and the protracted testimony and arguments engendered subsequent thereto are unimpressive. It was error at that time or not. It was not curable if it were error.

We think the only fair thing to do is to take evidence again as to damages, and if what plaintiff says in its brief about the incontestibility of its position, it should take comfort in the belief that no other verdict could eventuate.