OFFUTT, J., delivered the opinion of the Court.

On November 18th, 1935, Coolidge Snead, the infant son of Ella Snead, while crossing Presstman Street at its intersection with Mount Street in Baltimore City, was struck and injured by a taxicab, owned by Benjamin Mangione and operated by Gilbert R. Kines, his employee, under circumstances which appellees contended made the appellants liable to the infant plaintiff for the injuries he sustained as a result of the accident. To enforce that liability, Coolidge Snead, by his mother and next friend, brought an action in the Court of Common Pleas of Baltimore City against Mangione and Kines, and Ella Snead sued the same defendants in the same court for the loss of his services and for expenses incurred in treating his injuries. The two cases were tried together, the trial resulted in verdicts for both plaintiffs, and on each verdict a judgment was entered. These appeals are from those judgments.

Appellants concede that the evidence was legally sufficient to permit a recovery. They do not question the propriety of the court's rulings as to that issue, but they contend that the trial court (1) erred in refusing to instruct the jury that there was in the case no evidence legally sufficient to show that Coolidge Snead suffered any permanent injuries as a result of the accident, and (2) in admitting certain medical testimony concerning his injuries. These contentions are submitted by eleven exceptions, of which the first ten deal with rulings on the prayers.

The rulings on the admissibility of evidence all relate to an attempt on the part of the plaintiffs to show that, as a result of the injuries caused by the accident, the infant plaintiff's mind was permanently impaired and its development retarded. It appears without contradiction that he suffered no discoverable physical injuries of a permanent character, so that, unless the injuries did cause some permanent impairment of his mind, it necessarily follows that he was not entitled to recover compensation for permanent injuries. Whether they did have that effect was the real question in the case as it comes to this court. And, since the determination of that question depends in part at least upon the propriety of the rulings involved in the exceptions relating to the admissibility of evidence, they will be first considered.

In describing the accident, eyewitnesses said that the cab knocked the boy down and dragged him "about two doors"; that he made no outcry; that he lay in the street unconscious, and was bleeding from his mouth and nose. He was at once placed in a cab with his mother, who lived near the scene of the accident, and driven to the hospital. His mother said that just before they reached the hospital the boy asked where they were going; that he talked wildly and acted "like he was ready to run away"; that she took him by the arm and he walked into the hospital. She also said that on the night of the accident he had a "vomiting spell," and she sent for Dr. Hatcher, the family physician. In describing his condition before and after the accident she said: "Since the accident his conduct has been pretty good, but does not have good remembrance; that he gets into `mystery' all the time, that mystery means tearing up, destroying things; not long ago he set fire to the window curtains and they burned down; he daubs paint on the wall and when asked about it he would not know anything about it; was not like that before the accident; has not attended school since the accident; * * * that ever since the accident he has been wetting the bed, which he never did before; for at least three weeks after the accident he soiled his clothes during the day; did not do this before the accident; does not play with children like he did before; * * * he went to school in Accomac County, Virginia; his marks were good in school; he did not go as regularly as he should during the year he attended school in the country; he has been going to school in Baltimore since 1933; he was never involved in any other accident.

"Q. Will you tell his Honor and the gentlemen of the jury whether or not when you saw the boy he had any marks on his head? A. Yes, sir, he did. He had a scratch beside his nose and one over his forehead. She saw a hematoma, that anyone who looked at the boy could see his bleeding forehead and the side of his nose are hematoma because she saw it."

The first witness offered by the plaintiffs was one Henry Hammond, who described the accident and the boy's appearance and apparent injuries at that time. Following his testimony, Dr. Philip F. Lerner was called. He saw the boy for the first time in April, 1936, more than four months after the accident. After testifying that he had "obtained the history from the mother of an automobile accident in 1936" and that the boy had been under his care "until a week ago," he was asked to say "what if anything he found wrong" with the boy. Defendants objected to the question, but the court over that objection permitted the witness to answer it "subject to exception," and this testimony followed:

"A. I found that the child was backward, didn't answer questions which you would expect a normal child of his age to answer. (The Court) Doctor, suppose you give us any outward evidence of injury you found, if any at all. (The Witness) No physical signs, outside of hyperactive reflexes there were no external signs on the boy. That the boy was backward and did not react as a normal boy of nine years of age should react to questions, that he was unable to answer questions as to current events such as a normal child should, and did not respond as well as a normal child; that he rendered no treatment since the mother said he was under the care of a family physician."

The witness was then asked: "Doctor, assuming that this boy had had concussion of the brain on November 18th, 1935, would you say from your examination that the condition you have described would or would not be the result of a concussion of the brain?" An objection to that question was also overruled, and the witness allowed (also subject to exception) to answer, and he replied: "I think he had had a concussion; means disturbance in either behavior or mental condition; that he did not know the boy's condition so far as normality was concerned, prior to the accident; and that he had looked up his school record." Those rulings are the subject of the first, second, and third exceptions.

At the time those questions were asked, there was no evidence of any kind in the case that the boy had ever had a concussion of the brain, it did not appear that the witness knew anything of the physical injuries which he suffered as a result of the accident, except what he had learned from an undisclosed and unsworn statement by Ella Snead, and from a hospital record which contained these entries: "Provident Hospital, Accident Room Record No. 26158. Date 11/18/35, Name, Coolidge Snead. Address 1628 Presstman Street. Age 10. Colored. Sex, male. Occupation, school. Brought by G.R. Kines, Diamond Cab 968. Admitted to emergency room 2:40 P.M. Injury: Abrasion of left arm. Treatment: Cleansed, Scott's applied. Admitted? No. Sent to home. Condition good. Intoxicated, No." The witness did not testify that from any examination which he himself made that he was able to say either that the boy had had a concussion of the brain, or that his mental condition when the witness examined him was caused by the accident. Under those circumstances there was no basis for the assumption that the boy had suffered a concussion of the brain, or for allowing the witness to testify that the boy's mental condition was the result of such a concussion. He admitted that, while he had "looked up his school record," he knew nothing of the boy's condition "so far as normality was concerned" before the accident, and therefore had no factual basis for comparing his mental capacity at that time with his mental capacity when he came under the observation of the witness.

While he was not the "family physician" of the Sneads, Dr. Lerner may be classified as an attending physician of Coolidge Snead. As his attending physician, under the rule stated in Crockett v. Davis, 81 Md. 134, 149, 31 A. 710, he was qualified to state his opinion of the mental capacity of the boy when he came under his observation, without first stating the facts and circumstances upon which his opinion was based. Davidove v. Duvall, 160 Md. 345, 352, 153 A. 417, and cases there cited. But even so, if it definitely and clearly appeared that there was in fact no rational basis for the opinion, it had no probative force. Donnelly v. Donnelly, 156 Md. 81, 143 A. 648; Livingston v. Safe Dep. Tr. Co., 157 Md. 492, 146 A. 432; Davidove v. Duvall, supra.

It has been held ( Standard Accident Life Ins. Co. v. Wood, 116 Md. 575, 594, 82 A. 702) not to be reversible error to permit an expert witness to express an opinion based upon facts assumed but not in evidence when the question is asked, if such facts are later proved in the case. Generally, however, the evidential facts constituting the predicate should be in evidence when the question is asked, or the question should be accompanied by a definite offer to prove such facts. Jones on Evidence, sec. 371; 11 R.C.L. 585; 27 C.J. 713, 714; Robinson v. Jones, 105 Md. 62, 65 A. 814; Electric Light Co. v. Lusby, 100 Md. 634, 651, 60 A. 248. When the questions under consideration were asked, not only was there no evidence that the boy had suffered a concussion, but there was neither an offer to show that he had had a concussion, nor evidence subsequently given sufficient to prove such an injury. It is true counsel stated, "We will follow it up, sir," but that assurance, standing alone as it did, was so vague as to be meaningless.

The only evidence that the infant plaintiff had suffered a concussion of the brain was that of Dr. Bernard N. Hatcher, who testified that he "thought" the child had had a concussion. Dr. Hatcher was the family physician of the Sneads. He saw the child at his home on the night of the injury, and said that at that time he did not think there was much the matter with him, that he was conscious, knew Hatcher, but did not complain of any pain. He said that on the following day he found the bed in which the boy was lying wet from urine, but he did not apparently regard that as a symptom of concussion. He himself observed no recognized symptom of concussion of the brain, his opinion was, according to his own statement, "built upon the history" he obtained from the mother, and he said the only symptoms "that would possibly convince me were that he was stupid following the accident." He was unwilling to say definitely that the boy had had a concussion, so that his opinion was a mere conjecture, based in part upon the statement of the mother, which was hearsay, and the fact that the boy seemed stupid, and he assumed that he was stupid because he said he had no pain. Finally, when asked by the court: "Doctor, we get a little confused about this, you know. Your first testimony was, if you will recall, that when you first went in there you didn't consider there was anything much wrong with the boy? (The Witness) That's right. (The Court) Did you so testify? (The Witness) That's right. (The Court) Now, then, that brings us up to this: When was it that you began to have suspicions in your mind, suspicions that there was a possibility of concussion? (The Witness) That was at the first visit when I was there. (The Court) Well, then, how do you reconcile that — that's the thing I can't get clear; I always put myself in the position of the jury, and I like to clearly understand the testimony — if you thought there was nothing serious at the time of your first visit, how do you reconcile that with the thought there might be a concussion? He answered: "Well, I can answer that question this way: You see, sometimes concussions are so slight that they soon pass off and never give any after-effect whatsoever."

The mother, after Dr. Hatcher had testified, did say that the boy had had a "vomiting spell," but Dr. Hatcher did not know on the occasion of his first visit that the boy had had such an attack, and, although nausea was a recognized symptom of concussion of the brain, he did not inquire if he had been nauseated.

Such testimony was too indefinite and speculative to constitute the basis of a judicial conclusion, and it follows that the assumption in the question to Dr. Lerner that the boy had had a concussion was improper.

Nor did the fact that Dr. Lerner attended the boy during and after April, 1936, in the absence of facts, either observed by him or assumed, sufficient to justify the conclusion, qualify him to express an opinion as to the nature of injuries suffered in November, 1935. But the only tangible fact to support his conclusion that the boy had had a concussion of the brain, and that his present condition was a result of that injury, was that he was backward and did not respond to questions as a normal child of his age should, although he admitted that he knew nothing of the child's mental capacity prior to the accident. There was therefore error in these rulings.

In the examination of Dr. Hatcher, after he had testified that he had examined the boy on the night of the accident, he was asked, "Well, what did you find wrong with him, if anything?" The witness replied, "Well, the only thing I found wrong with him from an examination standpoint of view were these few scars on his face and arms; but I thought — " At that point counsel for defendants objected, the objection was overruled, and the witness continued, "But I thought he had had possibly a concussion of the brain from the accident." A motion to strike out that statement was also overruled, and the court said: "At what time? He had had the concussion of the brain at what time?" and Hatcher answered, "Well, I thought probably he had had it at the accident, you know, some time that afternoon." A motion to strike out that answer was also overruled. Those rulings are involved in the fifth exception. That exception refers not only to those rulings, but also to a ruling in reference to a question asked of another witness, but while the practice of embodying in one exception rulings on distinct question of law has been disapproved by this court ( Harris v. Hipsley, 122 Md. 418, 436, 89 A. 852; Pen Mar Co. v. Ashman, 152 Md. 273, 283, 136 A. 640; Winakur v. Sapourn, 156 Md. 662, 670, 145 A. 342), and is inconsistent with the provisions of Code, art. 5, sec. 12 (as amended by Laws 1927, ch. 224) since this case must be remanded for a new trial, it is deemed proper to review the rulings relating to the testimony of Dr. Hatcher which are included in the exception. As the attending and family physician he was qualified to describe the condition of the boy when he examined him ( Davidove v. Duvall, supra; Crockett v. Davis, supra), and the question first quoted above was proper.

The statement that the child had had "possibly" a concussion of the brain was too indefinite and should have been stricken out. The witness said later that he thought "probably he had had it at the accident." Such an opinion has been held sufficiently definite. 22 C.J. 641 et seq.; 11 R.C.L., 582. It was proper to permit Dr. Hatcher, as the attending physician, to express such an opinion without first stating its predicate, but, as it later appeared from his own admissions that there was no rational basis for it, while that fact did not affect the propriety of the rulings at the time they were made, it destroyed the probative force of the opinion.

Dr. Andrew C. Gillis, a specialist in nervous and mental diseases, was called as an expert witness by the plaintiffs. He had examined Coolidge Snead in his office the day before he testified, "got his history" from the mother, and had talked with Dr. Lerner. He testified that he had found the boy physically normal, but dull, stupid, with the mentality of a child of six or seven years, although he was apparently about ten years old, with over-active reflexes. He further testified that he had heard all of the testimony given by the mother. Upon those premises, after several false starts, the witness was asked this question: "Now, doctor, assuming her testimony to be true, and assuming that the boy was in some kind of an automobile accident and was in some way thrown to the ground and picked up and was unconscious and remained unconscious from that time and during the time he was taken over to East Baltimore until he got back to Provident Hospital, and that he had vomiting spells which the mother testified to this morning, assuming all that" — "hold that, will you, for just a minute, what I have put in that question? Because I was going to lead up to the findings of the doctor yesterday. And the mental findings are just dull, did you say, doctor" He said: "Yes, dull, and mentally retarded and from signs he has an intelligence of about seven." The question was then completed by adding this: "Assuming that he had vomiting spells as the mother testified to this morning, assuming all that, followed by periods of being kept in bed and then later was taken to the hospital where he was examined by the doctor who found that his reflexes were over-active, what would you say in your opinion to be the connection if any between the condition of the boy, what you heard the mother testify to this morning, and what you discovered yesterday afternoon?" An objection to that question was overruled, and the witness answered: "Assuming the evidence I have heard this morning to be true, and judging from my own examination of the boy, I would say that his present condition, his mental dullness, mental retardation, is the result of the injury which he received in November, 1935." A motion to strike out that answer was also overruled. Those rulings are the subject of the ninth exception.

The question was objectionable in form and substance, and should not have been allowed. It embodies an erroneous statement of fact that the boy was unconscious "until he got back to Provident Hospital," whereas it was uncontradicted that he recovered consciousness before he reached the hospital, and walked into it with his mother. Although it was directed to the ultimate issue in the case, whether the boy's mental condition was congenital or the result of the accident, it omitted any reference to his school record prior to the accident, it required the witness to resolve the conflict in the testimony as to the boy's age (his mother testified that he was nine years old at the time of the accident, November 18th, 1935, ten years old at the time of the trial, December 3rd, 1936, and that he was born in April, 1923), it states that he was taken to a hospital and examined by a physician who found his reflexes over-active, but failed to state that that examination was made over four months after the accident, and, although it attempted to summarize the testimony relating to the boy's physical injuries, it omitted facts essential to any rational conclusion upon the issue as to whether the boy's mental condition at the time of the trial was the result of the accident, such as the fact that Dr. Hatcher when he first examined the boy did not think "much was the matter with him," and that the hospital record failed to show any injury to the boy's head. The importance of the omission of any reference to his school record prior to the accident is shown by the following admission of the witness on cross-examination: "In other words, if he were not a normal child prior to the accident, and a dull child, and a stupid child, your answer would be exactly the reverse, would it not? A. Yes. If you were, of course, to show the intelligence as seven before the accident, naturally the accident would have nothing to do with it."

In Baltimore O.R. Co. v. Brooks, 158 Md. 149, 160, 148 A. 276, 280, the following statement in Northern Central R. Co. v. Green, 112 Md. 487, 505, 76 A. 90, is quoted with approval: "`A hypothetical question must embrace every material element of the hypothesis founded upon the evidence, and it must not import into the question any element not founded upon the evidence in the case. If it offends in either respect, it is defective, and it is error to permit such a question to be answered; and, if inadvertently admitted over an objection, it is error to refuse a motion to strike out the answer.' Berry Will Case, 93 Md. 560, 49 A. 401; Balto. O.R. Co. v. Dever, 112 Md. 296, 75 A. 352." See, also Gordon v. Opalecky, 152 Md. 536, 547, 137 A. 299. The question under consideration did not merely require the witness to express an opinion upon the effect of segregated facts in terms of physical or mental consequences, but to express an opinion upon the ultimate issue which the jury was required to determine, that is, the connection between the condition of the boy as he found it, and as the mother described it, and the accident. In such a case, when it is considered that the witness had no personal knowledge of the accident or the injuries, that he had never seen the boy until more than a year after the accident, the reason for the rule is apparent. The testimony of an expert witness without direct knowledge of the facts is only tolerated because it is assumed that because of his unusual skill and experience he will aid the court or jury to correctly interpret facts which might otherwise remain obscure. He cannot aid them, however, unless he is required to base his opinion upon the same facts and all the facts which the jury must consider in answering the same question.

Or, as stated in Hitchcock v. Burgett, 38 Mich. 501, 506, where a witness was asked: "`Assuming that the leg was in good condition prior to the accident, what should you say to be the cause of the difficulty, as you found the patient last?' The objection made to this was that the witness could not answer the question without assuming certain facts, and such facts it would be necessary to know in order to understand the value of the opinion. This objection was well taken. The witness was asked to give an opinion as to the cause of the particular difficulty under which the plaintiff was laboring at the time the witness made the examination in the fall of 1873. And it appeared also that the manner in which the plaintiff originally received the injury was known to the witness. Even in cases where experts are called upon to give an opinion based upon their own personal observation or examination, the facts upon which the opinion is founded must all be stated; otherwise the witness might be giving an opinion which would have great weight with the jury upon a state of facts very different from those found by them in the case on trial. The value of the opinion, in other words, must depend very largely upon the facts on which it is based."

While therefore it is permissible, in formulating the hypothesis of a question put to an expert witness, to base it in part upon testimony which the witness has heard and in part upon facts which he is asked to assume, that can only be done where there is no inherent conflict in the testimony he has heard, and no conflict between it and the stated facts. And whatever the form of the question may be, it may not omit any facts which may affect the conclusion sought by it.

Moreover, as stated in Quimby v. Greenhawk, 166 Md. 335, 338, 171 A. 59, 61: "since the inference or conclusion rests upon certain premises of fact, these premises must be true; and, in order that premises considered may be known and their truth or falsity be ascertained by the jury or the court, these premises must be expressly and particularly stated. For this reason the admissibility of a hypothetical question primarily depends upon whether it furnishes the tribunal with the means of knowing upon what premises of fact the conclusion is based."

Since the question under consideration violated these rules, the objection to it should have been sustained and the answer to it stricken out.

Following that testimony, Dr. Gillis was asked on redirect: "What was this boy's prospects for the future, Doctor?" An objection to that question was overruled and that ruling is the subject of the tenth exception. The question was improper. It may have referred to the boy's mental, social, or physical "prospects," it was based neither upon actual knowledge nor assumed facts, but was merely an invitation to idle and meaningless conjecture, and the objection should have been sustained.

The court at the conclusion of the whole case granted, at the instance of the plaintiffs, a prayer permitting the jury to award to the infant plaintiff damages based upon the hypothesis that he had suffered permanent injuries as a result of the accident, and it refused defendants' prayer that there was no evidence legally sufficient to show that he had suffered any such injuries. These rulings are the subject of the eleventh exception.

The defendants contend that the plaintiffs' prayer should have been refused because there was no evidence legally sufficient to show that the infant plaintiff suffered any permanent injuries, but, in the absence of a special exception to it on that ground, the ruling as to that prayer may be disregarded. Code, art. 5, sec. 10. The same question, however, was raised by the refusal of the defendants' sixteenth prayer and may therefore be considered.

There is in the record evidence that, following the accident, the infant plaintiff was stupid, erratic in his actions, that his memory was poor, that he got into "mystery" (obviously mischief was meant) all the time, that he wet his bed, and his mother testified that before the accident "the boy was normal," had a good memory, that his conduct at home and at school was good, that he did not wet his bed, and that after the accident he did not play with children as he did before he was hurt. The boy himself did not offer to testify and was not called as a witness, so that those statements are the only evidence of his condition before and after the accident, except certain school records, apparently offered by the plaintiffs, which showed that prior to the accident he had a mental age of seven or seven and a half years. His condition itself is not of such a character as to permit an inference that it is not likely to change, for, as stated by Chief Judge Bond in Cluster v. Upton, 165 Md. 566, 568, 168 A. 882, 883: "Prediction from present conditions of the conditions which will exist in the future may in some instances be made by a jury without the aid of opinions of experts in such matters. Common knowledge would tell them that a lost member is permanently lost ( Congoleum Nairn v. Brown, 158 Md. 285, 289, 148 A. 220, 67 A.L.R. 780), and there may be conditions which, while less obviously permanent in their nature, may still, according to common knowledge, be probably so, with such a degree of probability that an adjudication of permanency as a fact may be permissible without expert opinions. Washington, B. A. Electric R. Co. v. Cross, 142 Md. 500, 510, 121 A. 374; United Laundries Co. v. Bradford, 133 Md. 363, 367, 105 A. 303. There must, however, be a reliable basis for the adjudication of it as a fact, something beyond mere conjecture, or possibility; and the burden is upon the plaintiff to establish the fact by evidence sufficient to support the finding, if he intends to include permanent injury as an item in the ground of his recovery." See, also, Mt. Royal Cab Co. v. Dolan, 166 Md. 581, 585, 171 A. 854.

Apart from his condition itself, there was in the case no evidence legally sufficient to show that the boy's condition at the time of the trial was permanent in its nature, nor can such an inference be drawn from the condition itself. The mere fact that he was stupid, if that was true, was not in itself sufficient. If the stupidity was congenital, it may be assumed that he will continue to be stupid. If it resulted from his injuries, it may or may not be permanent, according to the physical effect of those injuries upon his mental and nervous organization. There was no admissible evidence as to what that effect was, nor indeed of any tangible, definite, or specific physical cause, existing at the time of the trial, for the mental dullness. In other words, before it can be said that the effect of an injury is permanent, it must appear that it is caused by some condition caused by the injury which is not likely to change, but such an inference cannot be drawn from the condition itself, when it is accompanied by no physical impairment or defect, is subjective, and offers no intrinsic indicia of its probable duration.

The defendants' sixteenth prayer should therefore have been granted.

The remaining exceptions were not pressed in this court and need not be considered, but for the errors indicated the judgment must be reversed.

Judgment reversed, and new trial awarded, with costs to the appellants.