taking in larceny must be a taking out of the possession of the owner, the inference from which would naturally be that when a thing is out of the owner's possession it cannot be stolen. We are then told, in order to avoid this conclusion, that a thing is always in its owner's possession; so that a box of plate at the bottom of the Thames, things of the existence of which the owner is not aware, as money vested in him as executor, and which without his knowledge is in the actual custody of another person, or a dead rabbit in his wood, are all in the owner's possession and capable of being taken out of it. This way of stating the matter makes the assertion that the taking in larceny must be a taking out of the owner's possession insignificant. If, from the nature of the case, every taking must be a taking out of the possession of the owner, it is impossible to see how the takings which do, differ from those which do not, constitute larceny. All men, being mortal, it is useless to define an Englishman as a mortal man living in England. However, though it is impossible either to justify the manner in which the word "possession" is used, or to free it entirely from the fictions with which it has been connected, it is, I think, not impossible to define it in such a manner as to express all the distinctions which it is intended to mark in language differing very slightly, if at all, from that which has generally been used upon the subject.

As I have shewn in the articles on theft, and in the notes upon them, there are five different ways in which theft can be committed viz:

1. By taking and carrying away goods which do not belong to the thief from any place where they happen to be.

2. By converting property entrusted by the owner to a

servant.

3. By obtaining the possession of property (as distinguished from the right of property) from the owner by fraud with intent to convert it.

4. By converting property given by the owner to the thief under a mistake.

5. By converting property bailed to the thief.

1 A. put 900 guineas in a secret drawer in a bureau and died. B., her son and executor, lent the bureau to his brother C., who took it to India, kept it there for several years, and brought it back. B. then sold it to D., who gave it to E. to repair, who found the money. This was held to be such a taking by E. out of the possession of A, as to constitute larceny. Cartwright v. Green, 8 Ves. 405.

It will be found upon consideration that the distinctions between these cases all arise out of the doctrine of possession, but it is, I think, less generally perceived that the important point is not the taking out of the possession of the owner, but the taking into the possession of the thief. The five cases in question may bo thus arranged:

In No. 1 (common larceny) the thief has neither the possession nor the custody of the stolen property at the time when the theft is committed, and it is immaterial whether the owner has it or not.

In No. 2 (larceny by a servant) the thief at the time of his offence may have either the custody or the possession. If he has the custody his offence is theft. If he has the possession his offence is embezzlement.

In No. 3. (larceny by trick) the thief obtains the possession by a mistake, caused by his own fraud.

In No. 4 (larceny by taking advantage of a mistake) the thief receives the possession by a mistake not caused by his own fraud.

In No. 5 (larceny by a bailee) the thief receives the possession under a contract of bailment.

Besides this view of the subject the doctrine of possession is important in relation to procedure, and in that case the matter to be considered is not the possession of the thief but the possession of the owner. It is necessary in indictments for theft that the ownership of the stolen property should be correctly stated, and as possession constitutes special ownership (at all events, as against a thief) it is important, with a view to this subject, to understand what possession implies.

Passing from the law upon this subject, let us examine the facts to which the law applies-the different relations which, as a fact, exist between men and things-in reference to the common use of language.

The most obvious case of possession is that of a person who holds something in his hand. But it must appear upon the slightest consideration that neither this nor any other physical act whatever can be accepted as more than an outward symbol of the state of things which the word denotes. Unless the article possessed is very small, part of it only can be held in the hand, trodden on by the foot, or so dealt with by any other part of the possessor's body as to exclude a similar dealing with it by others. It would however, I think, be felt by every one that neither actual bodily contact with an object, nor even exclusive bodily contact

with it, was essential to what, in the common use of language, is meant by possession. No one would think of using different words to express the relation of a man to a coin clenched in his fist, to a pocketbook in his pocket, to a portmanteau of which he carried one end and a railway porter the other, to a carriage in which he was seated whilst his servant was driving it, to a book on the shelves of his library, and to the plate in his pantry under the charge of his butler. He would, in the common use of language, be said to be in possession of all these things, and no one would feel any difficulty in perceiving the correctness of the expression even if it were added that he was not the owner of any one of them, that some had been lent, and others let to hire to him. On the other hand, any one but a lawyer would be surprised at the assertion that a man, whether the owner or not, was in possession of a watch which he had dropped into the Thames, of sheep which had been stolen from his field and driven to a distance by the thief, of a dead grouse which, having been wounded at a distance from his moor, had managed to reach it and die there without his knowledge or that of any other person.

The common feature of all the cases to which the word " possession" would obviously be applicable is easily recognised. It is to be found in the fact, that the person called the possessor has in each instance the power to act as if he were the owner of the thing possessed, whether he actually is the owner or not. Several of the illustrations given, however, shew that though this is one of the things which the word conveys, it is not the only thing conveyed by it. The butler in charge of the plate, the porter helping to carry the portmanteau, the coachman who is driving the coach, have the physical power of acting as the owner of those things as much as their master or employer. Indeed, in two of the three cases their physical control over the object is more direct than his. The difference is that the circumstances are such as to raise a presumption that their intention is to act under the orders of their superior, and that he (at least for the present) has no definite superior whose orders he intends to obey. Take, for instance, the case of a dinner party: there is no visible difference between the master of the house and his guests; each uses the article which he requires for the moment, and they are, from time to time, removed from place to place by the servants; as, however, the master retains throughout not merely the legal right to dispose of them absolutely, but the immediate means of enforcing that right

if from any strange circumstance it should become necessary to do so, the assertion that the plate is in his possession, and that his guests and servants have merely a permission to use it under his control, has a plain meaning: nor would that meaning be altered or obscured if the fact were added that the plate did not belong to the master of the house, but was hired by him for the occasion. Indeed, if he had stolen the plate, or received it knowing it to be stolen, the fact denoted by the word "possession" would remain. These illustrations, which might be multiplied to any extent, appear to me to shew clearly that possession means, in the common use of language, a power to act as the owner of a thing, coupled with a presumable intention to do so in case of need; and that the custody of a servant, or person, in a similar position, does not cxclude the possession by another, but differs from it in the presumable intention of the custodian to act under the orders of the possessor with reference to the thing possessed, and to give it up to him if he requires it. Thus far, I think, my definitions correspond with the common use of language, though of course popular language upon such a subject is not, nor is there any reason why it should be, minutely exact.1

I will now compare it with the way in which the word is used by legal authorities. I know of no set dissertations on the subject of the use of the word "possession" in English law like those which are to be found in abundance upon the corresponding word in Roman law. It would be an endless and a useless labour to go through the cases in which the word has been used, endless on account of their great number, useless because it is the characteristic of English judges to care little for technical niceties of language in comparison with substantial clearness of statement in reference to the actual matter in hand. Upon such a matter as this accordingly, it is better to consider the different authorities in groups than individually.

Possession (in reference to the subject of theft) is usually divided into two branches-actual possession and constructive possession. It seems to have been pretty generally assumed that the words

This view was suggested by a study of Savigny's Recht des Besitzes, which however, deals with many topics to which nothing in English law corresponds. Mr. George Long's article on "Possessio" in the Dictionary of Greek and Roman Antiquities contains the substance of Savigny in a very convenient form. Mr. Hunter's Roman Law, pp. 195-222, may also be consulted. It is a pleasure to me to refer to this excellent book.

"actual possession" were sufficiently plain for practical purposes without further explanation; but it would be easy to shew, by a multitude of cases, that actual possession differs from possession as I have defined it only in one point. It is usual to say that a thing in the possession of a servant on account of his master is only constructively in the possession of the master. But the expression "constructive possession" has another meaning besides this. As it was considered necessary that a thing stolen should be taken out of the possession of the owner, and as in very many instances goods are stolen which are not in any natural sense in the possession of any one whatever, it has become a maxim that goods are always in the possession of the owner; if not in his actual, then in his constructive possession, or, as it is sometimes called, in his legal possession.

Thus, constructive possession means:

1. The possession of goods in the custody of a servant on account of his master, and

2. The purely fictitious possession which the owner of goods is supposed to have, although they are in reality possessed by no one at all.

The phrase thus appears to me to be objectionable, not only because it is ambiguous, but because, in the first of its two senses, it conceals a truth, whilst in the second it needlessly conveys a false impression. The truth concealed is that a man may have, and may intend to use, the power implied in the word "possession," although he acts through a servant. The false impression conveyed is that things cannot be out of possession, or that if they are, they cannot be stolen.

I avoid this by abstaining altogether from the use of the expression" constructive possession." In "possession" Iinclude that which has to be exercised through a servant, and my language implies, that a person may commit theft on objects which are not in the possession of any one at the time of the theft. The existing law may by these means be expressed in well-recognised and established phraseology, without any resort to legal fictions.

The point upon which the most subtle questions as to possession arise, is the distinction between theft and embezzlement a perfectly useless distinction, no doubt, and one which the legislature has on two separate occasions vainly tried to abolish. So long, however, as it is allowed to exist, it is necessary to understand it.

I have already explained how a man may retain the possession of