DRUMAIOND, District Judge.
The defendant was indicted under the 21st and 22d sections of the post office act of Alarch 3, 1825 (4 Stat. 107-109), for stealing from the mail a packet containing a land warrant, and fifty dollars in bank notes. It appeared that the offense was committed near Dixon, on the 1st of August, 1850. The packet was mailed at Freeport on the 30th of July, addressed to Dixon. On the day the offense was committed, the defendant was arrested at the latter place, and a few days after-wards, a preliminary examination took place there before an officer. The defendant was present with his counsel, at the examination, during which one Hurlbut, since deceased,who had enclosed the land warrant and bank notes, and directed and posted' the letter, testified as a witness for the United States. Hurlbut was subjected—to use the language of the witnesses introduced here— to a long and tedious cross-examination toy the counsel of the defendant. An objection was taken by the counsel of the defendant at the trial in this court because witnesses were permitted to state' to the jury what Hurlbut had sworn to on the preliminary examination. This objection having been overruled, and " the defendant convicted by the jury, his [1133]*1133counsel has made a motion for a new trial, and it having been argued before me, I have examined the objection with more attention than I was able to bestow upon it at the trial. No question has been made of the power of the court to grant a new trial on the application of the defendant. Indeed, notwithstanding the doubts which have been thrown on that point heretofore, and the opinion expressed by Judge Story in U. S. v. Gibert [Case No. 15,204], I should have no hesitation in granting a new trial to a party who, I thought, was wrongfully convicted, more especially if it were caused, in any degree, by the erroneous ruling of the judge at the trial. See U. S. v. Harding [Id. 15,301]. In this ease there was other evidence, independent of the testimony of Hurlbut before the committing officer, which might have authorized the jury in finding their verdict; but there can bé no doubt that his testimony may have had much influence upon the jury, and, under the circumstances of this case, I should grant a new trial if I thought the testimony should have been excluded. After reflection, however, and all the examination I have been able to give to the subject, I am of the. opinion that the ruling at the trial was cor-| reet. '
The objection resolves itself into the two following propositions; First. The declarations of a deceased witness made at a former trial between the same parties, upon the same subject matter, can never be given in evidence in criminal eases. Secondly. If they can be, it is only when ' the persons who are called on to give the declarations of a deceased witness, can repeat the precise words of the witness, and it being admitted that that was not done here, the testimony ought to have been rejected. It is well known that there has long been a difference of opinion upon both these points. It is not controverted that the testimony of a deceased witness given at a former trial between the same parties, in the same issue, is admissible in civil eases. There seems no difference of opinion as to that. But some of the authorities &c., deny the application of the rule to criminal cases. A case which is generally cited as deciding that it does' not apply to criminal proceedings, is that of Sir John Fenwick in 1696. He was charged with being concerned in treasonable projects, but the witnesses who were expected to prove his guilt having left the country, there was no sufficient legal evidence to convict him of treason before the courts of law. The government resorted to a bill of attainder in' parliament A question arose whether the deposition of a person named Goodman, who was absent, taken before a justice of the peace when neither the defend-antnor his counsel was present, should be read as evidence? It was decided in the affirmative on the ground that the commons were not obliged to adhere to the rules established in Westminster Hall. During the discussions which took place in that case, it was said that such evidence could not be admitted in criminal cases in a court of law. Of . course it is clear that such testimony could not be admitted in a court of law; for, first, the witness was living; and, secondly,, the defendant had no opportunity of cross-examining him; and however the authorities may differ as to the first, they all agree as to the second point, that being an indispensable prerequisite to the introduction of the testimony. Mr. Parke relies upon this case for his assertion that there is a distinction between civil and criminal cases in this particular. The same opinion is expressed in Finn v. Com., 5 Rand. (Va.) 701, though the question there was, whether the testimony was admissible where the witness was absent. In Crary v. Sprague, 12 Wend. 41, Judge Nelson, who delivered the opinion of the court, while be admits it is questioned by high authority, and cites Hawkins and Parke, states his own opinion, that the testimony is admissible in criminal cases. In People v. Newman, 5 Hill, 295, while deciding that nothing but the death of the witness would authorize the admission of such testimony under the law of New York, the court say, if the rule were ' otherwise in civil cases, they thought it ought not to be applied to criminal proceedings. And they distinctly waive the question whether it would be allowed at all in criminal cases, if the witness were dead. Com. v. Richards, 18 Pick. 434, was a case very much like this. A witness who had testified against the defendant at a preliminary examination before a magistrate, having died, witnesses were called to state what the deceased witness had sworn at the examination. They were permitted to testify, and the case went to the supreme court of Massachusetts on this and another point. It was like this a case of an offence which was investigated in the first place before an officer, and the party was afterwards indicted for the same offense. The case was reversed, as we shall presently see, on another ground, but the court examined at some length the authorities for the purpose of ascertaining whether there is any distinction in this particular between civil and criminal proceedings, and came to the conclusion that there is none. In U. S. v. Wood [Case No. 15,756], which, like this, was a case of robbing the mail, though the testimony was rejected because the precise words could not be given, no allusion whatever is made to any difference between civil and criminal cases. Rex v. Joliffe, 4 Term R. 290, in which Lord Kenyon used these words in relation to the rule which has been since so often quoted, was a criminal information, and he speaks of no distinction.
Most of the modem elementary writers, Phillips, Starkie, Roscoe, and Greenleaf advert to the rule as one of general applica[1134]*1134tion in all cases. And Russell, particularly, in his valuable little treatise of the Law of Evidence, which he has' added to his work on Crimes, says expressly, the rule applies tQ criminal prosecutions. 2 Russ. Ev. 683. By the statutes of Philip and Mary, magistrates were directed and required to take the depositions of witnesses in certain criminal cases, and it has always been held, under these English statutes, that if the defendant were present at the taking of the deposition, and the witness were dead, it might be read on the trial as evidence. And yet there was nothing in the statutes from which it could be inferred that depositions were to be received as evidence.
Free access — add to your briefcase to read the full text and ask questions with AI
DRUMAIOND, District Judge.
The defendant was indicted under the 21st and 22d sections of the post office act of Alarch 3, 1825 (4 Stat. 107-109), for stealing from the mail a packet containing a land warrant, and fifty dollars in bank notes. It appeared that the offense was committed near Dixon, on the 1st of August, 1850. The packet was mailed at Freeport on the 30th of July, addressed to Dixon. On the day the offense was committed, the defendant was arrested at the latter place, and a few days after-wards, a preliminary examination took place there before an officer. The defendant was present with his counsel, at the examination, during which one Hurlbut, since deceased,who had enclosed the land warrant and bank notes, and directed and posted' the letter, testified as a witness for the United States. Hurlbut was subjected—to use the language of the witnesses introduced here— to a long and tedious cross-examination toy the counsel of the defendant. An objection was taken by the counsel of the defendant at the trial in this court because witnesses were permitted to state' to the jury what Hurlbut had sworn to on the preliminary examination. This objection having been overruled, and " the defendant convicted by the jury, his [1133]*1133counsel has made a motion for a new trial, and it having been argued before me, I have examined the objection with more attention than I was able to bestow upon it at the trial. No question has been made of the power of the court to grant a new trial on the application of the defendant. Indeed, notwithstanding the doubts which have been thrown on that point heretofore, and the opinion expressed by Judge Story in U. S. v. Gibert [Case No. 15,204], I should have no hesitation in granting a new trial to a party who, I thought, was wrongfully convicted, more especially if it were caused, in any degree, by the erroneous ruling of the judge at the trial. See U. S. v. Harding [Id. 15,301]. In this ease there was other evidence, independent of the testimony of Hurlbut before the committing officer, which might have authorized the jury in finding their verdict; but there can bé no doubt that his testimony may have had much influence upon the jury, and, under the circumstances of this case, I should grant a new trial if I thought the testimony should have been excluded. After reflection, however, and all the examination I have been able to give to the subject, I am of the. opinion that the ruling at the trial was cor-| reet. '
The objection resolves itself into the two following propositions; First. The declarations of a deceased witness made at a former trial between the same parties, upon the same subject matter, can never be given in evidence in criminal eases. Secondly. If they can be, it is only when ' the persons who are called on to give the declarations of a deceased witness, can repeat the precise words of the witness, and it being admitted that that was not done here, the testimony ought to have been rejected. It is well known that there has long been a difference of opinion upon both these points. It is not controverted that the testimony of a deceased witness given at a former trial between the same parties, in the same issue, is admissible in civil eases. There seems no difference of opinion as to that. But some of the authorities &c., deny the application of the rule to criminal cases. A case which is generally cited as deciding that it does' not apply to criminal proceedings, is that of Sir John Fenwick in 1696. He was charged with being concerned in treasonable projects, but the witnesses who were expected to prove his guilt having left the country, there was no sufficient legal evidence to convict him of treason before the courts of law. The government resorted to a bill of attainder in' parliament A question arose whether the deposition of a person named Goodman, who was absent, taken before a justice of the peace when neither the defend-antnor his counsel was present, should be read as evidence? It was decided in the affirmative on the ground that the commons were not obliged to adhere to the rules established in Westminster Hall. During the discussions which took place in that case, it was said that such evidence could not be admitted in criminal cases in a court of law. Of . course it is clear that such testimony could not be admitted in a court of law; for, first, the witness was living; and, secondly,, the defendant had no opportunity of cross-examining him; and however the authorities may differ as to the first, they all agree as to the second point, that being an indispensable prerequisite to the introduction of the testimony. Mr. Parke relies upon this case for his assertion that there is a distinction between civil and criminal cases in this particular. The same opinion is expressed in Finn v. Com., 5 Rand. (Va.) 701, though the question there was, whether the testimony was admissible where the witness was absent. In Crary v. Sprague, 12 Wend. 41, Judge Nelson, who delivered the opinion of the court, while be admits it is questioned by high authority, and cites Hawkins and Parke, states his own opinion, that the testimony is admissible in criminal cases. In People v. Newman, 5 Hill, 295, while deciding that nothing but the death of the witness would authorize the admission of such testimony under the law of New York, the court say, if the rule were ' otherwise in civil cases, they thought it ought not to be applied to criminal proceedings. And they distinctly waive the question whether it would be allowed at all in criminal cases, if the witness were dead. Com. v. Richards, 18 Pick. 434, was a case very much like this. A witness who had testified against the defendant at a preliminary examination before a magistrate, having died, witnesses were called to state what the deceased witness had sworn at the examination. They were permitted to testify, and the case went to the supreme court of Massachusetts on this and another point. It was like this a case of an offence which was investigated in the first place before an officer, and the party was afterwards indicted for the same offense. The case was reversed, as we shall presently see, on another ground, but the court examined at some length the authorities for the purpose of ascertaining whether there is any distinction in this particular between civil and criminal proceedings, and came to the conclusion that there is none. In U. S. v. Wood [Case No. 15,756], which, like this, was a case of robbing the mail, though the testimony was rejected because the precise words could not be given, no allusion whatever is made to any difference between civil and criminal cases. Rex v. Joliffe, 4 Term R. 290, in which Lord Kenyon used these words in relation to the rule which has been since so often quoted, was a criminal information, and he speaks of no distinction.
Most of the modem elementary writers, Phillips, Starkie, Roscoe, and Greenleaf advert to the rule as one of general applica[1134]*1134tion in all cases. And Russell, particularly, in his valuable little treatise of the Law of Evidence, which he has' added to his work on Crimes, says expressly, the rule applies tQ criminal prosecutions. 2 Russ. Ev. 683. By the statutes of Philip and Mary, magistrates were directed and required to take the depositions of witnesses in certain criminal cases, and it has always been held, under these English statutes, that if the defendant were present at the taking of the deposition, and the witness were dead, it might be read on the trial as evidence. And yet there was nothing in the statutes from which it could be inferred that depositions were to be received as evidence. But the law having sanctioned them, it seems they became admissible upon general principles, provided the defendant was present, had the liberty to cross-examine, and the witness was dead. See note to the case of Rex v. Smith, 2 Starkie, 208; 3 Eng. C. L. 316, § 6. These statutes require the magistrate to take the examination, as well of the prisoner as of the witnesses, in writing; still, if this was not done, it seems to have been the practice to admit the statements of the defendant and witnesses; though if the examination were reduced to writing, that must be produced. Rex v. Fearshire, 1 Leach, 240; Rex v. Jacobs, 1 Leach, 347; Rex v. Lasube, 2 Leach, 625. By the acts of congress, the officers before whom the persons charged with the commission of offense, are taken, have the right to allow bail. This implies the power to examine the facts of the case to ascertain whether the party shall be discharged, committed, or admitted to bail. The law does not require that the examination shall be reduced to writing. It was not done in this case. As a matter of practice, however, it is frequently done, and generally it is desirable that it should be. See the authorities collected in 2 Cow. & H. Notes to Phil. Ev. 571, or note 437, where the opinion is expressed that the statements of a deceased witness are admissible in a criminal the same as in a civil proceeding.
Without going into a further examination of the authorities in this branch of the case, it might, with some force, be argued that the weight of authority sustained the view of the court at the trial, but, however this may be, it seems plain that amidst so great a conflict of authorities, the court is at liberty to decide the question upon principle. Why should not the rule in civil and criminal proceedings be the same in this respect? The great object of all judicial investigation is to ascertain facts, and to do justice between the parties. In criminal cases, to shield the innocent, and punish the guilty. In accomplishing this, however, courts must act in conformity with some general rules founded in reason and experience. But after all our efforts we only make an approximation to this object. Many an innocent man has been and will be punished,—many a guilty one go free. If it be, on the whole, a sound rule to admit the declarations of a deceased witness, made on a former trial, in a case involving property or reputation, it is equally so in cases involving life and liberty. The ground upon which we proceed in each case is the presumption of the truth of the declarations, they being subjected to the tests which the law recognizes,—the presence of the accused, and the right of cross-examination. The admissibility of this species of evidence depends upon the necessity of the ease, and upon a well established exception to the rule which excludes hearsay, if, indeed, we may not in one sense, regard it as original testimony. We receive it because it comes up to one of the démands of the law; it is the best evidence which can be produced. Though the witness has been once confronted with the defendant, and, in-his presence, been sworn and cross-examined, it may be admitted, it is more satisfactory to have him again produced before a jury at a second trial, but being dead, it is impossible, and we resort to the next best source of truth,—his sworn statements already made. I think the law of evidence, as now administered, is quite stringent enough in excluding testimony, and I confess I feel a strong disposition to admit it in all cases where it can be done without violating any principle, or controverting any settled rule of law. It seems to me that to reject this kind of evidence, either in a civil or criminal case, is shutting out, without sufficient reason, one of the lights that should guide us in our judicial investigations.
The tendency of the courts in modern times in criminal cases is to afford the jury every opportunity that is consistent with the rules of law to determine the guilt or innocence of the accused, and I think they are peculiarly entitled to this sort of testimony, giving to it such weight as is proper under all the circumstances of the case. If the rule operates so as to expose guilt, it may protect innocence. ' There are anomalies enough in the law of evidence now without increasing them unnecessarily. In a criminal case involving life itself we admit, as testimony, declarations made by a person not under oath, and where the accused was not present. . And why? From the supposed necessity of the case, and because declarations made by a person under the danger of impending death, are regarded as if made upon oath in a court of justice.. We confine this rule of evidence to cases of homicide. But in the instance we are now considering, we have the sanction of the oath itself, administered by competent authority,—and the cross-examination of the witness,—the great test of truth, by the party; and there is thus every reasonable safeguard thrown around the claims of the public on the one hand, and the rights of the accused on the other.
Assuming, then, that the rule in civil and criminal cases ought to be, and is the same [1135]*1135in this particular, we come next to the other part of. the subject, and one not less controverted. I might, perhaps, put this portion of the case upon the ground that no objection was taken at the trial that the witness did not repeat the identical words of Hurlbut; but I do not choose to rest it there, having formed a decided opinion on this point. I am disposed to treat it as if the objection on that ground had been actually made, at the time.
The question is, whether a witness, when he is called upon to give the testimony of a. deceased witness at a former trial between the same parties, in order to render it admissible, must repeat the precise words used by the witness? or whether it is sufficient if he state the substance of what was sworn? Those who contend for the more narrow doctrine that the very words used must be repeated, refer to Lord Kenyon’s language in Hex v. Joliffie, already mentioned. He says, in referring to the instance of Lord Palmerston, in which, on all hands, it was agreed that this kind of testimony was admissible, “but as the person who wished to give Lord Palmerston’s evidence, could not undertake to give his words, but merely to swear to the effect of them, he was rejected.” “He ought,” said the same judge, in the case of Ennis v. Denisthome, referred to in 1 Phil. Ev. 219, “to recollect the very words, for the jury alone can judge of the effect of words.” Some judges have interpreted the language of Lord Kenyon, to mean that a person shall not state the effect, that is, give his own inferences as to what the deceased swore, but that he may state the very words substantially, that is, he must use the words of the witness, and not his own; and placing this construction on the rule, have agreed with him. Others, on the contrary, have denied his position altogether, and have insisted that the persons may give the substance or effect of what the witness swore; other judges, again, maintain that the meaning of Lord K. was, that the whole of the very words of the witness must be given. In restricting the rule within these limits, they admit that it is a virtual destruction of the rule itself.
In New l'ork and Massachusetts, the decisions ot whose courts we hold in high respect, it seems to be considered that the precise words must be used, and not the substance. This was the opinion expressed in Wilbur v. Selden, 6 Cow. 162, “The words must be given, and not what is supposed to be the substance of the testimony.” The case of Com. v. Richards, 18 Pick. 434, already cited, is a fair illustration of the effect of restricting the rule as contended for. A person, since dead, testified as a witness before the magistrate at the preliminary examination of the defendant, charged with a criminal offense. On the trial before the jury, two witnesses were sworn, and were permitted to repeat what the deceased witness had said before the magistrate. These witnesses did not re.peat the exact words used, but only the substance of them from recollection, aided by notes taken at the time. One of the witnesses said he was confident he stated the substantives and verbs correctly, but was not certain as to the prepositions and conjunctions. The supreme court reversed the case because these witnesses were permitted to state, under these circumstances, what the deceased witness had testified. The court, in giving its opinion, admits that such strictness will generally exclude that kind of evidence. They cite Rex v. Joliffe, U. S. v. Wood, and Wilbur v. Selden, authorities already mentioned. The question came up again in Massachusetts in Warren v. Nichols, 6 Metc. (Mass.) 201. In this case the witness said he could not give the words or precise language of the deceased witness, but he could give the substance of the testimony. The court differed in opinion, but a majority of the court adhered to the rule laid down íd Richards’ Case. I understand, however, the majority of the court to consider that the effect only of what the deceased witness said could be stated, and it was to be excluded on that ground. “As he could only give the substance and effect of the testimony, but not the language in which it was given,” it should be rejected. And yet, in another part of the opinion, the majority of the court say. “The witness must be able to state the language in which the testimony was given, substantially and in all material particulars,” that is, the witness must use the language of the deceased witness, and not his own; and it is to be inferred if they had supposed this had been done, they would have held it sufficient, though this principle is not easily reconciled with the decision in 18 Pick. 434, to enforce and strengthen which seems to have been the chief object of the opinion pronounced by the majority of the court in Warren v. Nichols. It would seem to be important in considering these two cases to bear in mind a distinction which many of the courts make, and which the court in Massachusetts seems not to have.had in view, between the effect of testimony, and the substance of it, or of the words used. In Indiana the same course has been followed. In Ephraims v. Murdock, 7 Blackf. 10, the court admitted there were conflicting decisions, but thought the weight of authority was in favor of the more stringent doctrine. And yet that court resisted an effort of counsel to make the same principle applicable to declarations in extremis. Where a person was indicted for murder, the witness could only repeat the substance of what was said, and not the exact words of the party, in extremis, and the court held this sufficient. Ward v. State, 8 Blackf. 101. On the other hand in Pennsylvania the courts held that it is sufficient for the witness to state the substance of what was sworn on the former trial. Cornell v. Green, 10 Serg. & R. 14. The court, while not very clearly distinguishing between the sub[1136]*1136stance and effect of evidence, enforces its view of the law in a very convincing manner. The judge says: “I cannot see why the same necessity which opens the way for secondary evidence of the very words of a deceased witness, should not open the way also for the substance of his testimony when his very words cannot be recollected; or discern the policy of a rule which should shut out the little light that is left, when it is all that is left, merely because it may not be sufficient to remove everything like obscurity.” And this was followed in subsequent cases. Smith v. Lane, 12 Serg. & R. 80; Chess v. Chess, 17 Serg. & R. 409. This construction of the rule seems to be approved by the courts of Maryland and Virginia, and by other states. Gildersleeve v. Caraway, 10 Ala. 260.
The question arose in a very late case in Ohio. Wagers v Dickey, 17 Ohio, 439. And the case is important as showing the rule adopted by the supreme court of that state, out of conflicting decisions upon the subject throughout the state. In that case the witness was permitted to state in substance what had been the testimony of the deceased witness; and the court says: “That if the strict rule is to be adopted, it amounts to a total interdict of a most important branch of secondary evidence, a branch of evidence especially important under our system of new trials.” And the judge who pronounced the opinion of the court in allusion to the strict construction of the rule uses this strong language: “The evils flowing from it would, in my judgment, were the law ever so well settled, justify a court in changing. the law.” In Illinois i* seems to be held sufficient for the words of the witness to be given substantially, and not the result of the evidence. Marshal v. Adams, 11 Ill. 37. All of our most approved writers on the Law of Evidence, Phillips, Starkie, and Greeuleaf, appear to prefer the most liberal construction of the rule. But it is needless to multiply authorities on the one or the other side of this vexed question. Most of the authorities (except the recent ones! can be found in 2 Cow. & H. Notes to Phil. Ev. 578, note 442. It might be matter of curiosity to examine how far the authorities could be reconciled by observing the distinction between the effect and the substance of the words used by the deceased witness,—a familiar illustration of which occurs in proving the words laid in a declaration in an action of slander. There it would not be sufficient to prove words to the effect of those used, nor even equivalent words, and yet it is enough to prove the words substantially as laid in the declaration. It is a singular practical commentary upon the rule that, in those eases where the courts maintain the strict doctrine, scarce an instance can be found where the evidence has come up to the' demands of the court, while it not unfrequently has happened where the courts have been more liberal in their construction of the rule, they have shown a strong disposition to restrain it within reasonable bounds. Watson v. Gilday, 11 Serg. & R. 337; Wolf v. Wyeth, 11 Serg & R. 149.
In our countrv new trials are so common that it often happens the testimony of a deceased witness is offered to be proved on a second trial. It is, therefore, a question of considerable practical importance. I have no hesitation in saying that I regard the more liberal view of the rule as being most in accordance with sound reason, and with the principles of evidence. All the analogies of the law are against the strict rule. Take the very common case of proof of verbal admissions of a party to the record. Do we reject the testimony-of the witness who proves these admissions because he cannot repeat the precise word? of the party? It is matter of every day’s practice to admit the evidence if he swears to the substance of what the party said. Another common case is that of lost writings. If evidence is in writing, it must, ordinarily, be produced, if in existence, and in the power of the party. If a written instrument is lost or destroyed, we resort to the next best evidence. Suppose that it is the memory of a witness who has seen and read the instrument. If he were introduced to a court, and should testify that he could state the substance of the words used in the instrument— that he was confident of the substantives and verbs, but not so certain of the prepositions and conjunctions (like the case in 18 Pick. 434), would his testimony be excluded because he could not repeat the exact words of the instrument? More especially, if, as he read it, he had taken notes of the contents, to assist his memory? I apprehend no court in this country would reject such testimony. In perjury we only require proof of the substance of the words upon which the perjury is assigned. We have already adverted to the case of slander, and to declarations, in extremis in cases of homicide. Take one of the few instances to which the majority of the court in Warren v. Nichols think the evidence must be confined,—a verbal notice of any fact given to a party. It is often necessary for us to prove such a notice. And would the fact that notice was given be rejected as evidence because the writer, .who was present, and who testifies to it, cannot repeat every preposition and conjunction made use of by the party Jn giving the notice? Most certainly even that court would say it is enough to prove it substantially. Besides, to admit the rule, and restrict it, as is sometimes done, makes it operate most unfairly. If the examination of a deceased witness has been short, and his words few, very possibly a person might recollect the precise words, but in a long and tedious examination, it would be morally impossible for any one to recollect all that the witness said; more particularly as the authorities agree that the witness must be able to state all that was said on the particular subject by the deceased witness, as web on the direct as cross-examination. [1137]*1137Now we know that the examination of a witness generally consists of interrogatory and answer, and if the precise words are to be given, then, tc be consistent, the question must be given as well as the reply. Take the case we are now considering as an illustration, and we will see the impracticability of this. In this case the witness, Hurlbut, was cross-examined at great length as to the identity of the bank notes, which he alleged were the same as those found on the defendant. Eve'y variety of question which the ingenuity of counsel could devise, was employed. Now, the witness introduced here, could testify to the substance of all that Hurlbut said in this examination, but to repeat the whole of what he said was clearly impossible; and ever will be in such cases unless it is all reduced to writing at the moment. I cannot persuade myself that it is reasonable to reject the substance of what he. said merely because we cannot have the very words. I understood the witness to say that he repeated substantially all that the deceased witness said, as well on the direct as on the cross-examination upon what -was the subject oí inquiry in this case, and therefore I think it was admissible. It seems as though the only safe course to pursue is to give this construction to the rule; or reject it altogether as some judges appear half inclined to do. To acknowledge the soundness of the rule one moment, and the next trammel it so as effectually to destroy it, seems like trifling. It is said that it "is a dangerous kind of evidence which, if resorted to¡ may be the means of doing injury. But the same may be said of every species of secondary evidence, and even of primary evidence itself. It is the lot of humanity. The main point is, Does the rule stand upon well established principles of evidence? Will it, on the whole, tend to promote the great ends of justice? The reason of the thing, and the authorities in the law have answered these questions in the affirmative. Believing that the rule rests upon clear principles, I think it should receive such an interpretation as shall make it of some practical utility, and not a dead letter.