general average contribution, and sometimes to denote such contribution itself. In order to avoid confusion, it would have been better to use the term "general average loss" when speaking of the former, and "general average contribution" when speaking of the latter. All losses which give a claim to general average contribution may be divided into two great classes: 1. Those which arise from sacrifices of part of the ship or part of the cargo purposely made in order to save the whole adventure from perishing. 2. Those which arise out of extraordinary expenses incurred for the joint benefit of both ship and cargo.

Losses of the first class are those which are alone mentioned in the text of that Rhodian Law which is generally regarded as the foundation of the whole doctrine of general average; but it is evident that expenses incurred by the owner of part of the adventure for the joint benefit of the whole give just as valid a claim to contribution in general average as any other species of loss intentionally incurred for the same purpose; and they have been accordingly admitted to give such a claim by the law and practice of all maritime states.

The only distinction between these two classes of losses is in the principles upon which they are contributed for, which, as we shall see in the sequel, vary in two cases; and upon this ground it becomes of practical importance to bear the distinction in mind.

A general average loss, therefore, may be defined to be a loss arising out of extraordinary sacrifices made, or extraordinary expenses incurred, for the joint benefit of ship and cargo.

In 1851 Mr. Lawrence R. Baily pub. General Average, and the Losses and Expenses therefrom. This is a useful little book, and has passed into a 2nd ed.

In 1857 Mr. Manley Hopkins pub. his Handbook of Average, which has passed through several eds. the 3rd in 1868. He treats the subject at once scholarly and popularly. Here are a few of his divisions. "Meaning of the word average-its origin.' "Damage to merchandize, or particular average." "General average-definition." Questionable subjects of general average.' "The contributories to general average." "Of average

connected with ins."

Mr. Hopkins moots a question which has often occurred to us, viz., "Whether it is desirable to retain one generic name, when the several species of claim embraced by it have become so individualized, and have such distinct characteristics?" But he also furnishes an answer which appears conclusive. "The word average is known throughout the world, and we therefore accept it in the sense in which it is ordinarily used." He then proceeds to see into what heads the word divides itself. These being:

1. General average. 2. Those charges and expenses which arise when a laden vessel puts into a port of distress, and which can be applied specifically to each of the interests-ship, cargo, and freight, frequently called particular or special charges. 3. The repairs of a ship rendered necessary by the perils of the seas. 4. Damages to merchandize by sea perils, including fire, plunder, etc. 5. Loss of freight by the decrease in quantity of the merchandize carried, by means of sea perils. 6. Partial loss of goods necessarily sold at an intermediate port, owing to sea perils. 7. Salvage loss of ship or of goods, recovered in part or in a deteriorated state, after being sunk, burnt, etc. 8. The constructive total loss of ship, goods, or freight. 9. The absolute loss or destruction of ship, goods, or freight. The last head is usually distinguished by lawyers from what they term "an average loss "; but on the grounds given above it is entitled to the same name as the rest.

We do not propose to pursue the subject in further detail here. It will be treated of in the course of this work under all its appropriate heads.

In 1860 there was pub. in Lond. a pamphlet bearing the title, What can they do with it? Remarks addressed to the Merchants and Insurers of Lond. on the proceedings affecting General Average, at the Social Science Congress lately held in Glasgow. By Amicus Curiæ. This refers to an attempt to estab. a General Average Law, to which end a species of congress had been affiliated upon the Social Science Asso. We shall speak of the matter under GENERAL AVERAGE LAW.

In 1867 Mr. Ernst Emil Wendt pub. Papers on Maritime Legislation, etc. The book is of a practical character, as might be expected from the large experience of the author. Its scope and objects may be gathered from one passage in its preface:

I am sorry to be obliged to record that, in spite of the repeated attempts made by the Associated Chambers of Commerce to prevail upon the Board of Trade to make the International General Average Rules, as adopted at the Congress of Delegates at York, in September, 1864, a subject of imperial legislation, nothing has been done.

In 1867 also, Mr. Francis B. Dixon pub. in New York A Practical Treatise on the Adjustment of General Average in the U.S. and other Countries, etc. This book is what it professes to be-essentially a practical treatise. We shall have occasion to speak of it more at large under GENERAL AVERAGE.

We may take this occasion to remark that several very learned treatises on the subject of average have appeared by American authors. We regret that from their not being ready of access here, we are compelled to pass over them in this art.

It was but natural that when, by means of marine ins., the losses which formerly fell upon the owners came to fall upon the underwriters, these latter should have also transferred to them all the benefits from salvage, etc., which the principle of average was calculated to confer upon them; but it prob. never was contemplated that the underwriters should find themselves in a worse position than the orig. owners. Yet it appears to us that the underwriter is frequently in a worse position. He must pay average as often as its cause arises during the voyage; and he must also pay total loss, so that averages and total loss combined might cause an underwriter (or other insurer) for £1000, to pay £1500 or even more. As to the owner, it would seem that the remedy is only against the goods to the extent of their value. As to the underwriter, the remedy is under his pol. We observe that the Charter of Ins. Co. of Copenhagen recognizes and provides for this.

In time of war the risk of average is hardly considered-the greater peril of capturewhich in many circumstances amounts to total loss, so far as the underwriter is concernedabsorbs the less; but in time of peace this risk is deemed of greater charge to insurers than even that of total losses. In France they have calculated the annual loss of shipping during 18 years of peace at one ship in 180; and the averages have been estimated at double that loss upon the like number.-Dict. du Citoyen.

The French Code de Commerce thus authoritatively defines average:—All extraordinary expenditures made for the ship and the goods, either conjointly or separately; all damage which happens to the ship and to the goods, from their loading and departure till their return and discharge, are reputed averages. [GENERAL AVERAGE.] [JETTISON.] [PARTICULAR AVERAGE.] [SALVAGE.]

AVERAGE POLICIES (FIRE).-These are pol. which contain what is technically called the "average clause "—and which imparts to such pol., in a limited degree, the principle of average, which is the leading feature of marine ins. These pol. are designed to cover merchandize in docks, warehouses, and bonded stores-called from the incidents of change, removal, and substitution, "FLOATING RISKS," in the U.S. "Blanket pol." Average pol. are indeed applicable to the largest mercantile transactions.

Over the greater part of the Continent of Europe-France, Belgium, Germany, and Russia-no fire pol. is issued without the provision of average or pro rata. The omission of it, in any branch of ins. (says Mr. Richard Atkins), throws so great an advantage on the side of the insured in the settlement of claims, that "it can be viewed in no other light than as equivalent to a very great, though unascertained reduction in the rate of prem." All the English pol. issued in Egypt, India, China, and Japan, contain the same conditions as to average as are contained in what are known as the Lond. or Liverpool "floating pol." In Batavia it is not so. In the U. K. an attempt has recently been made to apply the principle of average to farming stock ins. ; but private dwelling-houses are never ins. under such conditions here.

The cause of the omission of the pro rata principle in fire pol. of Gt. Brit., as contrasted with its general adoption in other parts of Europe, has given rise to some speculation. We think it may be discovered in this simple fact, that, whereas the first practice of fire ins. in many of the continental towns was by way of the Fire Casse, a sort of municipal fund, raised by a common assessment of all the house property of the place, to provide a fund for the protection of all-and therefore if any individuals failed to contribute their proper quota according to the value of their property, the pro rata principle would be applied in case of a loss. In Gt. Brit. the method adopted was entirely different. Fire ins. asso. were founded by a limited number of individuals, and they, as a matter of protection to themselves, limited the amount of ins. to a given sum-in the first instance £500. If a loss occurred, the insurers paid the damage incurred up to but not exceeding that particular sum, and there the transaction ended.

The next distinctive phase with us was the imposition of stamp duties upon the pol., which by the middle of the last century became in a great measure prohibitory of the spread of ins. to be followed in 1782 by the imposition of a duty, first of 1s. 6d. on every 100 insured; and increased from time to time until 1815, when it became 35. p. £100 or about double the prem. on first-class risks. Nor was this all: the restrictions of the Stamp Acts-framed with a view to the utmost exaction-made it very doubtful whether "Floating pol." were not beyond the scope of the law.

When, in spite of these and of many other restrictions, commerce yet began to assume stages of irrepressible advancement, then the number of fire offices began to multiplyso that every man could obtain, by means of separate pol. with the different offices, all the protection he required; and these new offices were strong and wealthy combinationscapable of undertaking individual specific risks of great magnitude. Whereas if the number of fire ins. offices had remained small, and their limits restricted, as in many of the European countries, the principle of average must have suggested itself.

At length it did do so, and then the difficulty of the legal restrictions of the Stamp Acts cropped up in full force. These were ultimately removed in 1828, by the 9 Geo. IV. c. 14, which contained the following provision:

III.-Provided always, and be it enacted that nothing in this Act contained shall extend or be construed to extend to prevent the ins. from loss or damage by fire collectively in one sum for the whole, any number of separate and distinct buildings, and the goods, wares, merchandize, or other movable property contained in any number of separate and distinct buildings, or lying or being in any number of separate and distinct places; provided that in the pol. whereby such ins. shall be made, there shall be contained a clause stipulating that in the event of any loss or damage by fire happening to such property, or to any part of such property thereby ins., the insurer or insurers in such pol. shall be liable to pay or make good such proportion only of the said loss or damage as the sum ins, shall bear to the whole collective value of the said property, at the time when such fire shall first break out or happen.

The clause, which was shaped under the authority of this Act, was, we believe, the following:

It is hereby declared and agreed, that in case of the property belonging to the insured in all the buildings, places, or limits herein described, shall, at the breaking out of any fire or fires, be collectively of greater value than the sum ins., then the co. shall pay and make good to the assured such a proportion only of the loss or damage as the sum ins. shall bear to the whole value of the property

aforesaid, at the time such fire or fires shall first happen. But it is at the same time declared and agreed, that if the within-mentioned assured shall at the time of any fire be ins. in this or any other office on any specific parcel of goods, or on goods in any specific building or buildings, place or places included in the term of this ins., this pol. shall not extend to cover the same, excepting only so far as relates to any excess of value beyond the amount of such specified ins. or insurances, which said excess is declared to be under the protection of the pol. and subject to average aforesaid.

The effect of this clause was and is, that if a merchant has goods of the value of £100,000 in one risk, and only insures £50,000 upon them, a fire occurring and destroying £40,000 worth of them, the merchant will receive from the ins. co. only £20,000-being the same proportion to the loss as the ins. bore to the whole value, viz., one-half.

It was soon found that the introduction of this clause into certain pol. while others were without it, might and did cause some difficulty in the adjustment of claims. An agreement was come to among the offices, as follows: When a fire occurred in any warehouse containing merchandize covered by "specified" pol.—that is, the old form of pol. against the new, or "floating" pol.-the specific pol. applying to that warehouse, was first to enter into settlement, and bear the whole of the loss, when that did not exceed the sum total insured. In cases where the sum of the specified pol. was not equal to the loss, the average pol. was to be brought forward in the second place, and to pay its share in the residue of the loss in the proportion that the sum ins. bore to the whole value of the goods covered in its range-after deducting the amount specially covered by the first pol.

This arrangement, however, did not go far enough. It did not reach the many complications resulting in cases of extensive destruction of mercantile property. Each office might have various pol. covering the same goods-either specific or floating; and many offices might jointly be interested in the loss under pol. of either, or both these classes. After much discussion, the practice settled down into the adoption of the following rule: That the floating pol. with the most limited ranges should be treated as specified pol. (after the actually specified pol. had taken up their proper burden)-in the order of their extent, and be called upon, in the first instance, in that order to make a settlement of the loss; while those of greater extent were only to be brought forward to cover any excess of loss which might be left after the prior settlement made by those of lesser range.—Atkins, The Average Clause, p. 16.

Matters went on under these arrangements until 1842, in which year some very destructive fires occurred. A great deal of difficulty arose in the adjustment of the losses, and without following these in detail, it will be enough for our present purposes to state that it was found that neither of the principles of adjustment already expounded would meet the case. The remedy devised was that known as the "Independent Liability Clause"-which abolished, in the case of floating pol., any priority of settlement on the ground of a greater or less range or extent of space covered by the terms of pol., and made the actual liability of the pol., independently estimated upon the average principle, the measure of the proportion which the office has to contribute to the loss. This alteration in the first instance was only made to apply to Liverpool floating pol.

In 1849 occurred the extended fire at the wool warehouses of Messrs. Gooch and Cousens. The salvage was unexpectedly large; the claimants for this very numerousabout 73 in all. Some were owners in their own right, and others simply as factors or agents. Some portions were ins. by specific, others by average pol. of every variety. Others again were wholly uninsured. Some of the proprietors by themselves or their agents put in their claim to distinct portions of the salvage-being able by the trademarks or otherwise to identify those portions. With the exception of these, the whole body of the owners were entitled to a pro rata share of the salvage, whether ins. or not. To add to the complication, many cases appeared where the expiration of the days of prompt, or the partial payment for goods sold, but not finally trans., and other similar transactions, made the legal ownership, to say the least, very doubtful. Hence the important determination was come to by the offices, that the warehouse-keeper's book should in all cases be taken as the test of ownership, for the purposes of settlement.

We next reach 1860, when the "Independent Liability Clause," adopted in 1843, but restricted to Liverpool up to this period, came into general adoption for average pol. of all ranges granted by the ins. offices of the U. K. Concurrently therewith the average clause-developed into "clauses" by the successive changes already indicated-came to be settled into the following shape, which they retain to the present day :

1. It is hereby declared and agreed, that whenever a sum insured is declared to be subject to the conditions of average, if the property so covered shall, at the breaking out of any fire, be collectively of greater value than the sum insured thereon, then the ins. co. shall pay or make good such a proportion only of the loss or damage as the sum so insured shall bear to the whole value of the said property at the time when such fire shall first happen.

2. But it is at the same time declared and agreed, that if any property included in such average shall, at the breaking out of any fire, be insured by any other pol., which, whether subject to average or not, shall apply to part only of the buildings or places or of the property to which such average extends, then this pol. shall not cover the same, excepting only as regards any excess of value beyond the amount of such more specific ins., which said excess is declared to be under the protection of this pol., and subject to average as aforesaid.

3. And it is further declared and agreed, that if the assured shall claim under this pol. for loss or damage to property embraced in the terms of any average pol. extending as well to other buildings or places, or to other property not included in the terms of this ins., and if at the breaking out of any fire there shall not be any property in such other buildings or places, or any such other property 16

VOL. I.

actually at risk to be protected by such pol., then, so far as regards the settlement of any claim under this pol., the terms and liability thereof shall be held to be concurrent, in all respects, with those of such other pol.

The fire offices have attempted more recently, and rather by a side wind as we consider, to extend the principle of average settlements by the introduction into specified pol. of the following, or a similar clause:

In case of the assured holding any pol. subject to average on the property covered by this assu., then this pol. shall be subject to average in like manner.

The effect of this clause, assuming it to be operative—which we consider somewhat doubtful-may and in many cases would be to cause the insured to receive less under the two pol. than he would under the former alone. We observe that Mr. Bunyon doubts the justice of the clause, "as it may be contended that the diligence of another set of insurers ought not to impart a fresh term into a prior contract to the prejudice of the assured." Mr. Hore says tersely, "There is reason to look upon this condition as illegal."

There can be no doubt that the introduction of the average clauses into fire pol. is greatly to the advantage of the ins. offices. Its direct effect is to prevent merchants and others from ins. below the value of their merchandise-on the speculation that being of necessity, or according to the custom of trade, stowed in various locations, the whole would not be destroyed by any one fire. Mr. Swinton Boult—and there can be no greater authority-said before the Select Parl. Committee on Fire Protection, 1867: “If the average clause were introduced into all fire pol., the effect would be two-fold. First it would increase the amount of ins., and next it would reduce the rate of prem. In France, as we have already stated, all fire ins. are made subject to average. The following is a translation of the clause adopted in the pol. :

If at the time of a fire the value of the objects covered by the pol. is found to exceed the total of the ins., the assured is considered as having remained his own insurer for that excess, and he is to bear in that character his proportion of the loss.

In Germany the principle of average in fire ins. is generally in use; but it is not at all an unusual circumstance, in regard to certain classes of risk, that the ins. offices should stipulate that the insured must keep at his own risk a certain proportion of the total value of the property insured. The following is the clause employed in the average pol. :

If in case of a fire the insured objects should exceed the sum insured, and they should be partly saved, the assured will be considered as self-insurer for the excess, and is to bear his share of the loss pro rata.

In the U.S. the practice is much like our own-the principle of average is only incorp. in the Floating, or as they are there termed "Blanket," pol. The clause there in use is to the following effect :

It is hereby agreed in case of loss under this pol, that the amount for which this co. shall be liable, shall be only such proportion of said loss as the amount hereby ins. bears to the whole value at risk.

It will be observed that all these clauses-France, Germany, and U.S.-are much more terse and compact than our own. The cause of this in the two first instances may be considered as due to the fact that, there being no other method of ins., the operation is in reality familiar to the mind. In the latter case the same reason cannot apply.

In 1856 Mr. George Coode, acting upon instructions from Sir George Cornewall Lewis, then Chancellor of the Exchequer, prepared a report on Fire Ins. Duties. Much of the information contained therein is of a very valuable character, but does not apply to this part of our subject. He does however deal with the subject of "average" in fire ins., with the view of estimating the influence of a general adoption of average fire pol. on the ann. amount of duty. Here again we need not follow him, for the duty has happily been abolished. In 1860 a revised ed. of his report was issued, and therein was contained the following among other obs. on average:

It is urged that owners, under the system of ins. in gross, unfairly pick out of their property the hazardous portions, giving to the insurers only the dangerous part to insure, withholding from them all share in the safe and better part. But this complaint appears to ignore the very nature and object of fire ins., which is not to insure from loss what is by its nature safe from destruction, but is to make men's fortunes secure, and their minds easy in respect of that which is in jeopardy of fire. An ins. to insure incombustibles from fire would be a self-contradiction. And as to the hardship on the insurers of having none but destructible property to insure, it must in reality amount to nothing, inasmuch as their prems., founded in experience of the losses incurred in this very system of insurance, are proportionally high, and are admitted to be at least fairly profitable, being prems. for entire ins. against loss, and would be just as much lower than they now are, if the system were altered so as to make the seeming ins. only a partial ins. against the loss, or in any degree an inoperative and nugatory ins. of indestructible property.

Another reason offered is that ins. on average, making the owner a partner in the risk, makes him interested in saving the insured property when in danger from fire, and so converts the owners themselves into an excellent fire police. But this reasoning somewhat conflicts with the general reasoning by which the change is justified, namely, that the tendency of the "average" system is to compel owners to ins. the whole amount of the property. Instead of operating to compel all persons to insure the whole of the property, in which case under the average system they would recover the whole of their loss, and be wholly uninterested pecuniarily in the salvage, this recommendation of the average system assumes that owners will under that system do as owners do under the system of ins. in gross-insure for a part only of the whole value of the property. It is clear that so far as the average plan induced owners to insure the whole value, this argument would entirely fail, while so far as it would hold good of partial ins. under that plan, it holds good of all under-insurances under

every possible system. In its principle, this is an argument against all ins. whatsoever, as diminishing the strength of the motives of the owner to save his property. But it is a fact testified generally by the officers of ins. cos., that all honest owners, under the present system of ins. in gross, do exert themselves to the utmost to save their property from fire, and the experience is almost invariable that honest owners fully insured, who can save nothing for themselves, constantly save all that it is possible to save. As to the fraudulent ins., who propose to make a gain by their apparent loss, they as a rule do ins. to more than the full value of their property. They are interested in its entire destruction, and in effacing every vestige which would afford a trace of their fraud. But this motive for ins. fully, and for effecting a total destruction, operates equally under the system of average ins. as of ins. in gross.

In 1866 Mr. Richard Atkins pub. The Average Clause: Hints on the Settlement of Claims for Losses by Fire under Mercantile Policies. From that work, which is regarded as of considerable authority, we have drawn some of the preceding facts.

In 1870 Mr. W. H. Hore, of the Liverpool, Lond. and Globe, pub. Remarks on the Apportionment of Fire Losses: illustrating by numerous examples the practice of offices in complicated Average and non-Average cases: and suggesting means by which the difficulties hitherto experienced in apportioning losses covered by non-concurrent pol. may be avoided in the future. [APPORTIONMENT OF FIRE LOSSES.]

In 1870 a resolution was arrived at by several of the F. offices most largely interested in Farming Stock Ins. that the "Average Clause" should be applied to these. [FARMING STOCK INS.]

Perhaps the only objection to the principle of Average Ins. is that unless property be valued at the time of ins., and from time to time afterwards, the policy-holder cannot be certain, in many cases, that he is, or continues fully insured. In the case of mercantile risks this objection does not apply-because every merchant has occasion in the course of his bus. to keep an approximate, if not an exact, value of his stock.

It will be obvious that in the case of loss under average pol., the office requires from the insured, in add. to the statement of loss made in the usual form, a valuation of the entire property under the protection of the pol. at the time of the fire. Full details of the requirements in this respect will be given under CLAIMS, FIRE.

While these pages have been in the hands of the printer, a case has arisen which may alter in some material respects the practice in regard to adjustments under average pol. In the grain trade it has hitherto been the custom of the wharfingers to effect the necessary ins. to cover stock during the time it remains in their custody. This they accomplish by means of average policies of limited range-that is, limited to the wharf or wharves belonging to them individually. In the case of the fire at the King and Queen Wharf, which happened in Dec. 1871, destroying grain to the value of £150,000 or upwards, it turned out that some of the owners of portions of this grain had, as a matter of extra precaution, included it in their general floating ins., under average pol. of a wider range than those of the wharfinger, viz., pol. covering their stock in these and any other wharves. Thus portions of the same grain were twice insured. A contention has arisen among the offices-for of course the losses to the owners of the grain have been paid-as to whether the last-named pol. ought not to rank pari passu with the pol. effected by the wharfinger: that is, take their proper share of the loss notwithstanding their wider range; and notwithstanding the rule which has been practised of ranking pol. of the more limited range first. A special case is to be submitted to the Judges, by whose decision the offices interested agree to be bound. The upshot of this question will probably be that the owners, and not the wharfingers, will in future insure their own stock. AVERAGE CALCULATIONS, STABILITY OF.-This subject will be discussed in our article on AVERAGES.

AVERAGE RISK.-It is well known to those who are really practically conversant with any branch of ins. bus., that, all other things being equal, the smaller the amount at risk in relation to any one contingency undertaken by the Co., the safer will be its bus. But it is also known that in many branches of the bus. the risks offered vary very much in amount-nowhere is this so marked as in fire ins. It therefore rests with the management to determine more especially what departure shall in any particular case be made from the average risks of the Co. This is a matter of vital consequence in the case of young offices. Various writers have discussed the matter in a theoretical point of view with great ability. We shall treat of the whole question under RISK, THEORY OF; and also to some extent under AVERAGES.

AVERAGE-STATERS.-See AVERAGE-Adjusters.

AVERAGES, DOCTRINE OF.-It may be said that the whole fabric of ins. rests on the practical truth involved in the principle or doctrine of averages. That casualties incident to persons or things will, under given circumstances, reproduce themselves, with but small variations, admits of no dispute, and indeed is never controverted. Hence then we know that a sufficient number of persons being exposed to risk, the casualties incident to life or limb will continually recur; and that even the physical conditions of the elements, which produce hailstorms in summer and frost and flood in winter, will repeat themselves. When we come to the number of houses to be destroyed by fire, or the number of ships cast away at sea, we reach events which are in some measure dependent on the human will; but even here averages, extending over longer periods, sustain themselves. Hence we place implicit confidence in the doctrine of averages. But inasmuch as no ins. asso. does or can obtain all the risks of any one class, and is subject, therefore, to the incidents