Sen, J.:— In this case there were four persons charged under section 395 of the Indian Penal Code, of these four, Madhu Mian has been acquitted the Judge accepting the unanimous verdict of the jury in his favour. As regards the other three accused, the jury by a majority of four to one, were of opinion that they were guilty under section 395 of the Indian Penal Code. The learned additional Sessions Judge took time to consider the verdict and finally passed an order accepting the verdict and convicting the accused. Bishu Tatwa, Bajit Mian and Baijnath Jolaha under the said section and sentencing them to rigorous imprisonment for four years each. This appeal is on behalf of two of the three accused so convicted and sentenced, namely, Bajit Mian and Baijnath Jolaha.

There are two grounds urged. The first is that having regard to the view of the learned additional Sessions Judge that the verdict was against the weight of evidence he should have submitted the case to the High Court under section 307 of the Code of Criminal Procedure. The second is that even if in the opinion of this court the learned additional Sessions Judge was not bound to make a reference under section 307, the order of conviction and sentence should be set aside inasmuch as the verdict of the jury was the result of misdirections on some of the most important issues in the case.

I propose to deal with the second ground first. It is necessary to consider some of the facts of the case in order to appreciate the arguments advanced under this head by learned Counsel appearing for the appellants. The dacoity in question is alleged to have taken place on the night of the 17th August, 1926, in village Basaita in the house of E one Mahadeo Gangota. Mahadeo Gangota was sleeping on the osara of the southern room of his house. After midnight he was roused by someone striking him with a lathi. Whereupon he entered the room opening on to the osara where his sons, Chedi, the first informant in the case, and Harihar were. The dacoits effected an entrance into the room and dragged out Mahadeo and Chedi into the angan. Mahadeo thereafter, entreated the dacoits to let him go offering to make over a hasuli to them. He went into the southern room, brought out a hasuli belonging to his wife, which had been hidden in the bedding, and made it over to the dacoits. Chedi in the meantime had effected his escape and ran into the fields. After the hasuli had been handed over, the dacoits began to intimidate Mahadeo's wife Mussammat Surji and tried to extract information from her as to where her valuables were hidden. They lighted a fire, threatened to burn her if she did not comply with their requisition, and even went to the length of scorching her leg in the fire they had lit for the purpose. Ultimately, when they discovered that there was not much else to be had from the house they retreated after taking such other small articles as they could get hold of. Chedi remained out in the fields the whole night. In the morning he sought out the village chaukidar and, with him and one Bhutan Gangota, proceeded to Kisimnganj thana where he arrived at 6 p.m on the 19th August, 1926. It happened, however, that there was no one at the thana who had power to institute a case. There was only a, literate constable in charge of the thana who made an entry in the station diary and sent on Chedi and his companions with a copy of this entry to village Lachmipur. They reached Lachmipur at 11 p.m, and there the writer head constable drew up a regular first information report. A great deal of the argument advanced in this court turns upon this first information. Chedi while lodging the first information report stated that he himself did not know who the dacoits were. All that he could say was that his father had said that there were eight dacoits in all, that they were speaking Urdu and that all of them were young. He added,

“I could not see on account of darkness if they had applied anything to their faces.”

Later on, however, there appears to have been some development inasmuch as Mussammat Surji is alleged to have identified Bajit, Baijnath and also Bishu Tatwa, the last named with particular reference to his gold stopped teeth. It may be noted that in view of this evidence, given in the course of the trial the question naturally arose as to whether if Mussammat Surji had so definitely recognised the three accused, their names would not have been stated in the first information report lodged by Chedi. It was suggested on behalf of the prosecution that it was just possible that Chedi had not come back home from the fields before going to the police station to lodge the first information. This theory, however, as appears from the charge to the jury administered by the learned additional Sessions Judge, could not stand for a moment. In point of fact Chedi, it is suggested in the charge, must have come home and had a consultation with his father and mother and seen what damage the dacoits had done to the house and the yard, and having so apprised himself of the main facts and circumstances connected with the dacoity he must have proceeded to the police station to lodge the first information. The, learned Additional Sessions Judge makes special reference to the fact that Chedi, having been away in the fields most of the time when the dacoits were in the house, could not possibly have given the details of the occurrence and of the damage done to the house, etc., of his own knowledge had he not returned to the house first and then proceeded to the police station. In all these matters it is perfectly clear that the learned Additional Sessions Judge made a full and fair and adequate charge to the E jury finally conveying his opinion to the jury that

“It is impossible to believe that Chedi had no talk with Mahadeo and. Mussammat Surji before he left the village. He must have had, because he himself was not present throughout the occurrence and many of the details given in the first information report must, and could only, have been derived from his parents.”

Learned Counsel, appearing for the appellants, contends, however, that this dogmatic expression of the Judge's own opinion would rather take away from its weight and convincingness; the better course, and in fact the only proper course, for the learned Additional Sessions Judge was to place the direct evidence on the subject. He points out a passage from the deposition of Mussammat Surji in which she said,

“Chedi also came back. One Bhutan of lower Lowngaon came with him. Chedi went to the thana. It was about midday when he left.”

He also points out a passage from the deposition of Chedi in which he said,

“I have known Bisu Tatwa for 10 years or so. I have known Bachhu and Baijnath for about 5 years. I have known Bajit Mian for the last 4 years I went to see them occasionally at the at or on the road * * * I recognised them by their voices. They were Bajir, Bisu, Baijnath and Madhu, the accused. I asked why they were maltreating us but they went on shouting stop, you sala or we will kill you”.

On the strength of these two passages, one in the deposition of Chedi and the other in that of Mussammat Surji, learned Counsel contends that if the Judge had put these statements before the jury which constituted the direct evidence in the case, instead of giving his inference in a dogmatic manner as above mentioned, the jury would have been in a position from such direct evidence to come to right conclusions on the question as to whether the omission of the names from the first information would not sufficiently establish the innocence of the accused in regard to any part in the dacoity in question. In other words, instead of telling the jury that it was impossible to believe that Chedi had not returned home before proceeding to the thana the learned Additional Sessions Judge should have given the jury Mussammat Surji's evidence to the effect that Chedi did come as a matter of fact and that he, therefore, had sufficient opportunity to know the names of the accused. Further, he argues that the evidence of Chedi himself that he had actually recognised the accused, if put to the jury, would have led them at once to view with very great suspicion the first information report the absence from which of the names of the accused would in that case be perfectly unaccountable.

A third contention which he raises is that the learned Sessions Judge was equally dogmatic in that portion of the charge where he discussed the question as to whether such a dacoity did actually take place or not. He expresses the definite opinion that a dacoity had in fact occurred at the place and on the date as alleged. The learned Additional Sessions Judge observed,

“It is impossible to believe that his son would, if no dacoity had taken place at all, have lodged the first information an which the present case was instituted.”

He also said:

“It seems to me impossible not to believe that a dacoity was actually committed and that a reasonably veracious, account has been given of what the dacoits did at Mahadeo's house.”

After having so delivered the charge on the subject of the occurrence itself the learned Sessions Judge proceeded to remark that the real question at issue, however, was whether the four persons in the dock were among the dacoits. This according to the learned Counsel for the defence is denying the jury the right to form their own judgment on the question as to whether a dacoity did take place as alleged. Having failed to infect them with his own belief so dogmatically expressed, it was not to be wondered that the learned Additional Sessions Judge was eventually surprised at the verdict of the jury, against all the accused.

These are the main objections on the ground of misdirection to the jury. Barring these features which in his opinion are objectionable though perfectly unintentional, learned Counsel concedes that the charge is, as it reads, a charge for acquittal and the learned Additional Sessions Judge has clearly and unambiguously expressed his opinion that on the evidence adduced in the case it would hardly be possible to bring the charge home to the accused. It is difficult to judge from the heads of charge to the jury which must necessarily be an abstract, and not a verbatim report of the summing up of the Court, as to whether the passages in question in the deposition of Mussammat Surji and Chedi were or were not placed before the jury [Eknath Sahay v. King-Emperor]. But the learned Additional Sessions Judge, in my opinion, has not merely in a dogmatic manner expressed his belief in the matter of Chedi's return home and also in the matter of Chedi's supplying himself with material facts for the first information from his parents. In one portion of his summing up he observes,

“It is not reasonable to suppose and there is evidence to the contrary, that Mussammat Surji did not communicate to her son before he left the village all that she knew of the occurrence and in particular of the identity of the decoits.”

The “evidence to the contrary” may certainly refer to the evidence of Mussammat Surji that Chedi came back home before going to the thana. I am, therefore, constrained to think that there is no substance in the contention that by omitting to refer to any material evidence in the case the Court prevented the jury from arriving at the right conclusion. As I have already observed the summing up of the learned additional Sessions Judge from beginning to end is so clearly in favour of acquittal that it would be straining too much to say that the jury did not receive adequate assistance from the Court. It is only when the non-direction is such that there are grounds for thinking that the jury by reason of it may have been put on the wrong track and made to arrive at a wrong conclusion that such non-direction can amount to a misdirection fail to see any such non-direction in the summing up.

Nor do think that there is much force in the contention put forward that the opinion expressed by the learned Additional Sessions Judge that a dacoity in fact took place as distinguished from his opinion that it was highly doubtful whether the accused took part in it would amount to a misdirection. If any thing, it would, in my opinion, more pointedly refer to the innocence, of the accused notwithstanding the fact, as it seems to have been established by the evidence, that a dacoity had taken place. The result, therefore, is that I do not think that there is any occasion for this Court to interfere on the ground of misdirection.

It remains now to consider the other contention, namely, that in the circumstances of this case the learned Additional Sessions Judge should have made a reference under section 307 of the Code of Criminal Procedure. There are three things which appear clearly from the provisions of section 307 and from the rulings thereon. First, it is entirely within the discretion of the Judge as to whether he should make a reference or not. Secondly, a reference is called for when the judge “is clearly of opinion that it is necessary for the ends of justice to submit the case in respect of such accused person to the High Court” and he shall in that event submit the case accordingly, recording the grounds, of his opinion. Thirdly, the Judge must disagree with the verdict of the jurors or of a majority of them. In the present case the learned Additional Sessions Judge took time to consider whether the case should be referred to the High Court under section 307 of the Code, of Criminal Procedure. He did so, as he states in his order, on account of the fact that at the time he had a suspicion that the methods adopted by the police had been reprehensible. Subsequently, he made a more careful study of the police diaries and found that his suspicion was not well grounded. Accordingly, he modified his original view and felt satisfied that the appellants had taken part in the offence charged against them. In these circumstances he recorded an order as follows:—

“Although therefore, I still think that possibly the conviction of Bajit and Baijnath is against the weight of evidence on the record, I have modified my original view and an satisfied that these persons took part in the crime. There is thus no reason why a reference should be made to the High Court. I accept the verdict and convict the accused.”

It is to be noted that although he was originally minded to dissent from the verdict, at the moment of the order he was certainly not in a position to dissent from it. On the contrary, he had fully satisfied himself as to the guilt of the accused. I fail to see how in these circumstances it could be a fit case for reference under section 307. Learned Counsel for the appellants, however, urges that inasmuch as there was still some doubt in his mind as to whether the conviction of Baijit and Baijnath was not against the weight of evidence on the record the only course left to him was to submit the case to the High Court. Reliance is placed on the case of Queen-Empress v. Guruvadu in which it is laid down that the discretionary power to refer cases under section 307 should always be exercised when the Judge thinks that the verdict is not supported by the evidence inasmuch as a failure to do so results in the conviction of persons on evidence as to the sufficiency of which the Court is doubtful. The facts of that case, however, were different. The learned Sessions Judge in that case being doubtful as to whether the verdict was justified by the evidence thought that it was not incumbent upon him to send the case to the High Court observing that “probably there would be an appeal.” The High Court in that case pointed out that the learned Sessions Judge should have remembered that there was no appeal on the facts when once the trying Court had accepted the verdict and based its conviction thereon; thus there was no justification for the failure to submit the case to the High Court on the ground that probably an appeal would be preferred. Another case upon which strong reliance has been placed is the case of Saroda Charan Mistri v. King-Emperor. In that case the jury brought in an unanimous verdict of guilty under section 302 of the Indian Penal Code in respect of one of the accused and under section 328 of the Indian Penal Code in respect of the other accused. The learned Judge observed that there was no justification for the differentiation of the cases of the two accused and that if the accused were at all guilty they would then both be certainly guilty of murder. On this point, therefore, there appeared to be a clear, full and complete dissent on the part of the learned Judge from the verdict of the jury. The other point involved in the case was as to whether or not the two accused knew that the drug they were administering to the deceased was poison. They had been given that drug by a third person as a medicine. The case of the defence was that the accused did not know while they administered the drug that they were administering poison. The jury unhesitatingly rejected the case of the defence as to such absence of knowledge. The learned Judge was inclined to think that he would rather give the accused the benefit of the doubt. But ultimately when passing the order he observed that considering that the jury had unanimously brought in a verdict of guilty he did not think he was justified in pressing “the slight doubts” that still remained in his mind to the extent of referring the case to the High Court under section 307 of the Criminal Procedure Code. With reference to the last mentioned point the High Court came to the conclusion that the disagreement of the learned Judge with the jury was not such a complete dissent as to lead him to consider it necessary for the ends of justice to submit the case to the High Court. But with reference to the former the High Court held that there was such a clear and complete dissent as to the liability of the two accused under the same section that he was obviously unable to do justice to the accused by accepting the verdict and, therefore, there was no option left to him but to refer the case to the High Court under the provisions of section 307. The present case bears some resemblance to the case of King Emperor v. Saroda Charan Mistri in so far as it relates to “the slight doubts” that still remained in the mind of the Court. Such doubts have been expressed by the learned Judge in the words:

“Possibly, the conviction of Bajit and Baijnath is against the weight of evidence on the record.”

In the next sentence, however, he observes,

“I have modified my original view and an satisfied that these persons took part in the crime.”

This takes the case completely beyond the scope of the provisions of section 307 of the Code of Criminal Procedure. But it cannot be said that he disagrees with the verdict. Nor that he is clearly of opinion that it is necessary for the ends of justice to submit the case to the High Court. This principle has been affirmed in the later case Eran Khan v. Emperor which lays down that “section 307 quite clearly gives to the Judge a discretion in the matter, and it is only when he is clearly of opinion that it is necessary for the ends of justice to submit the case to the High Court that he shall so submit it. If he is not clearly of that opinion his failure to submit the case is not a subject for interference by this Court on appeal.”

In this connection it would be well to remember the principle laid down in one of the earliest cases of jury trial in India: “If we are to interfere in every case of doubt, in every case in which it may with propriety be said that the evidence would have warranted a different verdict, then we must hold real trial by jury is absolutely at an end and that the verdict of the jury is of no more weight than the opinion of assessors,” Queen v. Sham Bagdee. In the present case it was unexpected that the jury should have returned a verdict of guilty by a majority of four to one. But the jurors are entitled to their own view of the case and the rule of law is not to disturb their verdict unless it be for special reasons and under special circumstances. The principle underlying that rule is well expressed as follows: “We adhere generally to the principle notwithstanding our large discretionary powers first, on the constitutional ground of taking as little as possible out of the hands to which it has been primarily assigned by the Legislature, and secondly, because any undue interference may tend to diminish the responsibility which it is desirable that a jury should cherish” Reg v. Ghanderav Bajirav.

The appeal must be dismissed and the convictions and sentences affirmed.

Allanson, J.:— I agree.