thereby substituted for A., for (1): C., being substituted for B., who again was substituted for A., was impliedly substituted for A. Hence the rule-substitutus substituto censetur substitutus instituto.

In this case, if B. had not been substituted for A., the jus accrescendi would have produced the same result. For C., the substitute of B., by taking B.'s share, was entitled, by the jus accrescendi, to the vacant share of A. Prior to Justinian, it was not so. At that time, the jus accrescendi was restricted by the caducary laws (the lex Papia Poppaa), which conferred either upon the hæres with children, or upon the public treasury (ærarium), the shares caduca or quasi-caduca, i.e., all that had become vacant, either before the testator's death, or between his death and the opening of the testament (vide tit. Legacies). As we have said, substitutio prevented a bequest becoming caducum. Moreover, even after Justinian's time there was an advantage attached to substitutio, in case there were three instituti: thus, when for the first you substitute the second, and for the second a fourth, non-institutus, such fourth, in case the two first instituti failed, received together with the share of the second, the whole of the first: whereas, had there been no implied substitutio, the share of the first, when vacant, must have been divided by the jus accrescendi between the third institutus and the fourth, viz., the substitutus of the second (2).

TITLE XVI.-OF THE SUBSTITUTIO PUPILLARIS.

Pr. Substitutio pupillaris is the appointment or institution of a hæres made by a father for a son under his power (3), in case the son should die a pupillus. For no impubes could make a will; and therefore, to prevent intestacies, a custom was introduced allowing a pater-familias to make a will for those children, who, at his death, would be pupilli, i.e., sui juris, and impuberes (4). But no pater-familias had this privilege unless he himself had a testament (§ 5) or was a soldier.

§ 2. The substitutio pupillaris presupposed two testaments: the father's and the son's, or at least a double testament operating on two

no

(1) § 3. Sine distinctione, i. e., difference was made in consequence of one substitution preceding the other; nor does it matter whether A. or B. is the first to die or to refuse to accept the hæreditas. Such distinctions were abolished by Severus and Antoninus.

(2) Shortly: A., B., C., instituti; D., non-institutus: B. for A., D. for B. substituti then by substitutio D. takes the

share of A. and of B. by the jus accrescendi D. takes the share of B. and half the share of A., C. taking the other half.

(3) At the time when the father makes the substitutio, and at the time when he dies.

(4) The pater-familias may appoint a substitutus pupillaris, whenever he may nominate a testamentary tutor (B. 1, t. 13).

subject-matters, since it disposed of two hæreditates. If, however, the father used one instrument for himself and another for his son, the father's will was required to precede the son's, but when he used only one instrument for both, no order was prescribed.

§ 4. A pater-familias might appoint a substitutus not only for children in the first degree but for the more remote issue, provided that on his death they would be sui juris. Moreover, he might do so for posthumous children.

§ 3. In order to prevent any danger to a son by his father publicly nominating a substitutus for him, who would be interested in the son's death whilst a pupillus, certain precautions were taken. The father might leave open the first part of the will, disposing of his own hæreditas, and tie or seal up the other part, nominating the substitutus, desiring at the same time that the scals should not be broken before the death or full age of the pupillus.

§ 4. A pater-familias was not bound to institute as hæres the filiusfamilias for whom he appointed a substitutus pupillaris. On the contrary he might appoint such a substitutus for his issue, whether he instituted or disinherited them. When the son was appointed hæres, his substitutus pupillaris was regarded as a substitutus vulgaris; and vice versa, the substitutus vulgaris was regarded as his substitutus pupillaris, unless the testator expressed a different intention.

§ 4. Upon the substitutus pupillaris devolved all the goods vested in the pupillus by succession, gift, or otherwise. But if the fatheradrogator made the substitution, it affected no goods but those received from or through him (D. 28, t. 6, 10, § 6). The other goods of the adrogatus went to his legal successor or to the substitutus appointed by his natural father.

§ 6. The father might appoint a substitutus pupillaris for each of his children. And he might also appoint one for the last which should die a pupil. In the first case no child died intestate; in the second the order of legal succession was retained and the child who died last a pupil was the only one which had a testamentary hæres.

Pr. § 7. There were various forms of appointing substituti. They might be named, e.g., Titius, be my son's hares: or generally, Whoever shall be my hæres shall be my son's hæres. In this last case the hæreditas of the pupil passed to those instituti of the father, who became in fact hæredes; it passed to them, however, not in equal shares, as it would have done if they had been individually named substituti, but according to the proportion of their shares in the father's hæreditas (1).

(1) If a slave, appointed hæres by a father, and substitutus pupillaris of the son, became free after acquiring for his master the father's hæreditas, but before the son's

death under age, such slave, and not his former master, was the substitutus: for the benefit of the substitutio was personal (D. 28, t. 6. 3).

§ 8. Substitutio pupillaris was ended: 1. By the pupil reaching puberty, fourteen years old in the case of boys, and twelve in the case of girls; for he could then make a testament for himself. 2. When the father's testament was annulled; for the substitutio pupillaris was incident to it. 3. When the child died before the father; for then the right to make a will had never vested in it. 4. By any diminutio capitis suffered by the pupil, either before or after the testator's death ; because, in the first case, the testator was divested of the patria potestas; and, in the second case, the pupil must either have ceased to be a citizen or to be sui juris, and therefore to have the right to make a will (1).

§ 9. If a stranger or a son of full age was appointed hæres by a testator, he could not provide for the case of these persons becoming hæredes, and then dying within a certain period, by means of a substitution, putting other persons in their place as hæredes. All the testator could do was to bind, by fideicommissum (trust), the stranger or the son of full age to give the whole or part of his hæreditas to a third party. But there was a great difference between such an institution by means of a trust and the substitutio pupillaris; for in the case of a trust institution, the testator disposed not of the hæreditas of the institutus, but of his own hæreditas, and therefore the proposed transferee of the hæreditas not being the hæres of the trustee, or person charged with the fideicommissum, succeeded to none of the trustee's goods, and received nothing more from him than he had received from the testator.

§ 1. There were, however, certain puberes for whom a testament might be made. Thus Justinian allowed ancestors of lunatics being puberes to appoint substituti, who should succeed them if they died before recovering their reason. This was the substitutio exemplaris or quasi-pupillaris.

§ 1. The substitutio pupillaris differs from substitutio exemplaris in several particulars. 1. The substitutio exemplaris might be made not only by the pater-familias, but by any ancestor in the male or female line. 2. In the substitutio exemplaris the substituti must be selected from certain persons (certas personas, § 1), i. e., from the children of the lunatic; and, in default thereof, from his brothers; so that the right of selection was not absolute, as in the substitutio pupillaris,

(1) Observe, however, that in case the pupillus was adrogatus, and died under age, his goods were to be given back to those who would have had them if there had been no adrogatio, and consequently to the substitutus of the adrogatus appointed by his natural father. But then it was not directly by virtue of the substitution

that the substitutus sued, but utiliter by
virtue of the stipulatio by which the adro-
gator was bound to restore the goods of the
adrogatus. For it is to be observed that
we are discussing only persons adrogated
whilst under age. In such cases the .
adrogator was bound to give a security
(I. B. t. 11, § 3).

TITLE XVII.-HOW TESTAMENTS ARE AVOIDED.

Pr. Though a testament were made with the required solemnity (t. 10), by one who had the right and the power to make it (t. 12); and though the sui hæredes (t. 13) had not been omitted and a fit hæres (t. 14), had been instituted, still such a testament did not always take effect. It was valid in its creation (jure factum); but it might be avoided (rumpatur), or become ineffectual (irritum).

§ 5. The words illegal (injustum), ruptum, and irritum are often used synonymously; but as it is always better to distinguish each thing by its own name (§ 5), a testament is injustum when void in its creation ; ruptum when it is valid in its creation, but afterwards avoided through some cause unconnected with the status and capacity of the testator; irritum when it is avoided by some change occurring in the status and rights of the testator (§ 4).

§ 1. A testament is said to be ruptum: 1. By the unexpected addition or birth of any hares who has not been instituted or legally disinherited (1). 2. By a subsequent testament, capable of giving a new hæres to the testator. 3. By the testator legally revoking one testament without making another.

§ 2. We say that the testament was ruptum by a subsequent testament, capable of giving a hæres to the testator (§ 1), because, if the second testament was regularly made (jure facto), and valid in its creation, it avoided the first. If, therefore, causes arising after the date of the second testament made it ineffectual: if, e.g., the hæres appointed by it refused, or died before the testator: or if the condition on which his appointment as hæres depended, was not fulfilled, the pater-familias in all such cases died intestate, for his first testament ceased the moment the second came into existence, and the second supplied no hæres.

The second testament destroyed the first without express revocation : a person could not die with two testaments, because each must contain the appointment of a hares; but that disposed of the whole rights of a person, and therefore could no more be made twice, than the same thing can be given twice. Now, as the intention of a testator might vary until he died, of two testaments, the later in date must destroy the other.

§ 3. If the second testament expressly confirmed the first, then seeing

(1) T. 13, ante. Observe that the breaking of the testament, by a child being adopted into the family, could not be pre-vented by an exhæredatio, unless in the case of a child which was originally emancipated by the same person who afterwards

adopted it. In this case the child was only returning to his original family. But such breaking might be prevented by instituting him, for though you could not take from a stranger the title of hæres, you might bestow it on him by testament.

that it was impossible for the same person to have two testaments, the first was held to be extinguished. But in order to give the testator's will as much effect as possible, the first testament was held to be a codicil (t. 25), and the confirmation contained in the second was construed as a trust, binding the hæres to restore the hæreditas to those instituti under the first, minus the fourth, which every hæres charged with a trust is entitled to by the extended operation of the Falcidian Law (t. 23).

§ 3. The first testament was invalidated, when by the second the institutio was confined to a particular article (ex certis rebus): because it was a rule to disregard that part of an institution which confined it to a particular article.

§ 7. But the second testament, if imperfect, did not revoke the first : because though it certainly indicated a change of intention, still that was not enough to destroy an institution: there must be a legal revocation (1).

However a testator might revoke his testament without making a second; by tearing it up, or otherwise destroying it with the intention of dying intestate (D. 28, t. 4, 1.); or by declaring his change of will publicly, or before three witnesses. Such declaration, however, did not instantly revoke the testament, for the testament was required to be ten years old (C. vi. 23. 27).

§ 4. A testament was ineffectual (irritum), when the testator suffered a diminutio capitis.

§ 6. Even if the testator regained his original status before his death, the testament did not revive by the Civil Law (2); but if the testament was then sealed by seven witnesses, the Prætor granted to those instituted thereby the possession of goods secundum tabulas, provided the testator had the testamenti-factio, both when it was made and when he died, without regard to the intervening period.

(1) The Senate decreed this on the proposal of Pertinax. In the same senatusconsultum he declared-1. That he would not accept an hæreditas if it was given (litis causa) from hatred to those with whom the testator had had disputes, in order that they might have a powerful adversary to contend with. 2. That he would not validate testaments to which he had been instituted, because they were null. 3. That he would not be hæres if that office had been conferred upon him by mere word of

mouth (ex nuda voce), or even by writing, unless in the proper form. For, said the emperor, though we are above the laws, we put ourselves willingly under them.

(2) For the testator who changed his status lost his original person, his former patrimonium. If he again became sui juris he had a new person, a new patrimonium, to which his former testament did not strictly apply. As to Postliminium, B. 2, t. 12; as to soldiers, B. 2, t. 11.