2. In certain other instances of acquiring community property, a title or a chain of title apparently perfect and valid on the record must necessarily be entirely nugatory and void, or else its actual validity must be established by facts completely outside of the record and resting upon the memory and verbal evidence of witnesses. Community property consisting of land is acquired by and conveyed to a wife during the marriage; she afterwards conveys the same parcel of land, and subsequently by successive deeds it is conveyed to A. All these deeds put on record make a perfect chain of title to the wife, and a perfect chain of record title from her to A., who on the records appears to be clothed with a perfectly valid legal title as the present owner. The records would not necessarily disclose the fact that the land was community property, and ordinarily this fact would not affirmatively appear on the records. So far as they are concerned, they show an unbroken chain of conveyances vesting a perfect title in A. And yet, because the property is community, the deed from the wife is void, and the whole chain of title from her down to A. is simply nugatory. The fact that the land is community property may be, and ordinarily must be, established by extrinsic evidence; and thus a chain of title appearing to be perfect and valid on the records, is entirely defeated and overthrown by facts outside of the record, which may be, and generally must be, proved by verbal testimony. A person intending to purchase the land from A., upon making a search of the records would find a title seeming to be without a flaw; but if, relying upon the record title, he should purchase, he might discover that he had been misled and that the title was completely worthless.

But this is not all. Even if in a case similar to the last, the land acquired by and conveyed to the wife was in reality her separate estate;— if, for example, it was paid for by money which came to her as a legacy, or a distributive share, or as the proceeds of other separate property of hers sold, or in other like manner, but the deed to her, as is ordinarily the case, failed to show such fact, but merely showed a conveyance to her for a certain purchase price or pecuniary consideration; then, as has

been previously shown, a strong presumption arises that the land is community property. If under such circumstances the wife conveys the land, and the title comes through successive deeds to A., this chain of title from her to A. would be presumptively void. Although on the record the deraignment of title is complete and unbroken, apparently vesting a perfect ownership in A., yet the validity of his title would really depend upon facts wholly dehors the record,-facts resting perhaps upon the memory of witnesses,-namely the fact that the land was conveyed to the wife during her marriage, and the fact that the purchase price paid by her was money belonging to her separate estate. An intended purchaser from A., although a search of the records disclosed no flaw in the title, could not safely rely upon the apparently perfect record title alone; a knowledge of the extrinsic facts not shown by the record would be essential to his complete security.

The same conclusions are equally true, if, under like circumstances, a wife should give a mortgage upon land conveyed to her which was community property, either actually or presumptively. Even though the mortgage so given had been regularly foreclosed, and the land sold at the foreclosure sale, and the purchaser's deed thereof had been executed, delivered, and recorded; the title so acquired, and appearing on the records to be perfect, might be entirely nugatory.

3. There is still a third case. In certain other instances of acquiring community property, there may be two conflicting chains of title to the same parcel of land on the record, one unbroken and appearing to be perfect, and the other interrupted and broken, and appearing to be imperfect, and yet the latter, by the aid of extrinsic facts outside the record, may be shown to be the true, valid and perfect title. A parcel of land being community property is acquired by and conveyed to a wife during the marriage. She afterwards gives a deed of the same parcel, and by successive deeds it is finally conveyed to A. The records of these deeds, not disclosing the fact that the land was community property, would show an apparently unbroken and perfect chain of title through the wife and ending in A., and vesting a valid title in him. On the other hand, the husband, after the conveyance to the wife, gives a deed of the same parcel, and through successive deeds from his grantee, it is finally conveyed to B. This second chain of title appears on the record to be broken and imperfect. The deraignment of the title on the record from the original source down through the wife to B., is interrupted and incomplete, since nothing appears on the record by which the husband's title is connected with that of the wife, nothing showing how the title apparently vested in the wife has passed to the husband. So far as the record discloses, the husband and wife are mere strangers, and the chain of title on the record commencing with him is separate from and uncon

nected with the original source of title. Looking alone at the record, the chain of title from the original source through the wife to A., is complete and perfect, and that from the husband to B. is imperfect and invalid. And yet from the facts outside of the record it is plain that the conveyances from the husband form the only true chain of title, conneeted with the original source, and vesting the real ownership in B., while the conveyances from the wife are nugatory and A.'s title is wholly invalid.

A similar uncertainty or imperfection might arise from the records even though the land conveyed to the wife was her own separate property. Land is conveyed to a wife during her marriage which is really her separate property, but as in the case supposed under the preceding head, the deed to her fails to show this fact in any manner by any of its recitals, but is merely a conveyance expressed to be for a pecuniary consideration paid by her. She afterwards conveys it, and by subsequent successive deeds it comes to A. The husband also assumes to convey it, and by subsequent successive deeds from his grantee it comes to B. Although the chain of title on the record from the original source through the wife to A. is apparently complete and perfect, yet the validity of A.'s title might depend entirely upon facts outside of the record, resting in the memory of witnesses, and to be established by their verbal testimony. The fact that the conveyance was made to the wife during her marriage would raise a presumption that the land was community property; and this presumption could only be overcome, and B.'s title derived from the husband could only be defeated, by proof of extrinsic facts showing that the land was the wife's separate property,--such as that the purchase price was money coming to her by bequest or gift, or otherwise belonging to the separate estate. These facts, being entirely outside of the record, would generally rest in the recollection of witnesses and be proved by their parol testimony.

The same results would also follow if the conflicting titles in such a case from the husband and the wife were derived through mortgages given by either or both of them, instead of deeds.

It must not be supposed that the illustrations which I have given are impossible or improbable cases The facts of actual cases which have been decided by the court and reported, furnish ample precedent and authority for every circumstance and every disposition of property described in the foregoing illustrations. That husbands can, and must, convey the community property held in the name of their wives, is a well established truth, and not a mere supposition. Nothing is more likely than that a wife should sometimes, and not infrequently, give deeds or mortgages of community property which had been acquired by or conveyed to her, and of which she appeared to hold the title. Even though

this might not be done secretly and with the intention of violating her husband's rights, or of defrauding her grantee or mortgagee, it might well be done through a mistaken notion on her part that she had the power to convey, or that the land itself was her separate property. Although the wife had thus attempted to convey or mortgage the land, still the husband would retain his full and sole power of disposition over it, and nothing could prevent or hinder his giving a deed or mortgage of it, notwithstanding the prior and really nugatory transfer by his wife. In this manner the apparently valid, but actually void, chains of title on the record from the wife, and the two conflicting chains of title from the husband and the wife, which I have described, are not only possible but probable; indeed it may be said that they are certain to occur. Nor is the other uncertainty which I described, resulting from a conveyance of the wife's own separate property, at all improbable. When land which is really her separate property is conveyed to a wife during her marriage, but the deed to her fails to disclose the facts showing it to be her separate estate, so that the legal presumption of its being community property arises, the husband will sometimes take advantage of this presumplion, will treat it as community property, and will assume the sole power of disposition over it by conveyance or mortgage. This will even be done though the husband knows that he is violating the rights of his wife, and defrauding his grantee or mortgagee. The reports contain too many similar cases to warrant any one in denying the probability of such a transaction. We have seen that a mistaken claim that land, really the wife's separate property, was community property, may arise and be maintained in good faith after the death of both the spouses.

It has thus been shown,—and, it would seem, beyond the possibility of doubt, that from the peculiar doctrines concerning the acquisition, disposition, and transfer of community property, uncertainties, discrepancies, and contradictions may, and must, appear on the records of titles; chains of title apparently good on the record may be defeated by extrinsic facts; and titles appearing to be perfect on the record may require to be sustained by extrinsic facts. In short, that reliance cannot always be placed upon a seeming perfect record, but resort must sometimes be had to facts outside the record-facts resting in the memory of persons, and to be established by their parol testimony. It must naturally follow that the inconvenience resulting from these imperfections of record titles will be felt to a greater degree after a lapse of time, when death has removed the witnesses upon whose personal knowledge and evidence the facts involving the validity of such titles must rest.

5th. It is natural to enquire, in this connection, whether there are any means or methods by which these uncertainties and other imper

fections may be wholly or partly prevented or removed? There are, as it seems to me, certain very simple and practicable methods, which if generally adopted would go far to obviate, even if they did not entirely remove and overcome, the various uncertainties, discrepancies and contradictions among record titles hereinbefore described. Undoubtedly, these methods have already occurred to and been practised by various members of the bar, but the suggestion, now made, that they should be adopted, recommended and followed on every occasion by all lawyers, may perhaps be deemed important and worthy of careful consideration. The settled rules are, as we have seen, that property, for example land, conveyed to the wife during the marriage by a deed expressing the conveyance to be for a valuable consideration, - that is, reciting or stating a purchase price, -is presumed to be community property; but if the property so conveyed is her separate estate, this fact may be shown and the presumption rebutted, by proof that the purchase price paid was itself her separate property, as, that it was a legacy, distributive share, gift, proceeds of the sale of other separate property, and the like. I would therefore suggest as a universal rule of great practical benefit, that whenever land is conveyed by a third person to a wife which is her separate property, recitals should always be inserted in the deed describing fully and in detail the nature and source of the consideration or purchase price and thus clearly showing that the consideration or price was itself also her separate property. Thus, the recital should state that the purchase price was money received by her as a legacy, or as a distributive share of her father's estate, or as a gift from her husband, or as proceeds of the sale of other separate property specified, or that it was other separate property of hers given in exchange, or whatever the real fact might be. Recitals of material facts in a deed or other instrument are expressly made by the code of civil procedure to be conclusive upon the parties and their successors in interest.' These recitals would of course bind the wife and her grantees and heirs claiming the land as separate property; and as the husband and his grantees claiming it to be community property, must derive their title through and from this deed to the wife, they would also be bound. The presumption that the land was community property arising from the mere fact of the purchase during the marriage, would thus be completely overcome and destroyed by matter appearing on the face of the deed itself. By this simple expedient all doubt as to the land being the wife's separate property would be forever removed, and the necessity of a subsequent resort to parol evidence in order to establish its character as her separate property, would be wholly obviated. It is possible that a mere statement in the deed that the land conveyed to her was her separate estate, 1 Code of Civil Procedure, § 1,962.