Chapter XIV: Part 14

This tableau, in which the player only wins 9 out of 29 coups--or, say, one in three--may be said to be far out of proportion, as the player is "entitled" to win as many coups as he loses (leaving zero out of the question). Let it be noted at this point that zero does not affect a system played on the even chances in any degree whatsoever. Any system worthy of the name can withstand zero, even two or three zeros. It is the Bank's limit, and the limit alone, that proves the downfall of all systems. To resume. Of course a player "ought" to win two coups out of four, and so he will as a rule, and systems are devised so that a player may be a winner, even if he loses three and four times as many coups as he wins. A glance at those figures not yet erased in the example quoted will show that had the punter not been debarred from staking, owing to the Bank's limit, with three successive wins he would have got all his money back and been ten points to the good on the whole transaction, and {459} still have only won twelve times against the Bank's twenty. What no system, played with a Martingale, has yet been able to accomplish, is to prevent the stakes becoming colossal when the series of losses turn up in some particular sequence or disposition.

The best method to keep the stakes within reasonable limits, and to guard against arriving at the Bank's maximum on an adverse run, is to employ a varying unit. Thus after a net loss of so many single units, operations are re-started with a double unit; if an equal number of double units are lost, the play is re-started with a triple unit, and so on; the same unit being employed until all previous losses have been retrieved, and a gain of one "single" unit made.

A "_Montant et demontant_" system can be played very easily in this manner, by increasing the unit employed after each complete loss of ten units--_e.g._ after a loss of 10 single units, the system is started afresh with a double unit; when 10 double units have been lost, or a net loss of 30, the system is started afresh with a 3 unit stake, and so on.

This system may be varied by changing the unit after successive losses of 10, 20, 30, 40, &c., and by staking sufficient to show a net win of the amount of the unit employed. Thus when playing with a double unit, to try and win 2; or if playing with a unit of 5, to try and win 5 units net.

Every system has its Waterloo--it will succeed for days, possibly weeks, and small gains be made; but finally the occasion must and will arrive when all previous profits and the system player's capital will be swamped. At the end of this article will be found a scheme devised by the writer whereby the punter puts himself into the position of the Banker as nearly {460} as possible, and consequently is enabled to win such vast stakes as are lost by a system player in the ordinary course, when that particular sequence of events occur which demolishes his system.

Here is an example of a "_Montant et demontant_" played in the usual method, and played with an increasing unit after each net loss of 10 units. The player is supposed to stake on the Red throughout; and the dot indicates which colour wins.

+-------------------+-------------------+------------------------------+
| Ordinary | A varying Unit | |
| Method. | employed. | Remarks. |
+-------------------+-------------------+ |
| R. | B. | Net | R. | B. | Net | |
| | | +or- | | | +or- | |
+------+-----+------+------+-----+------+------------------------------+
| 1 | . | -1 | 1 | . | -1 | |
| 2 | . | -3 | 2 | . | -3 | |
| 3 | . | -6 | 3 | . | -6 | |
| 4 | . | -10 | 4 | . | -10 | Having lost 10 single units, |
| | | +------+-----+------+ the system is re-started |
| 5 | . | -15 | 2 | . | -12 | with a double unit. |
| 6 | . | -21 | 4 | . | -16 | |
| 7 . | | -14 | 6 . | | -10 | |
| 8 | . | -22 | 8 | . | -18 | |
| 9 . | | -13 | 10 . | | -8 | |
| 10 | . | -23 | 9 | . | -17 | As the object is to be +1, |
| 11 | . | -34 | 11 | . | -28 | 9 is a sufficiently high |
| | | | | | | stake. |
| 12 | . | -46 | 2 | . | -30 | As not more than 30 may |
| | | +------+-----+------+ be lost while employing |
| 13 | . | -59 | 3 | . | -33 | a double unit, 2 is the |
| 14 . | | -45 | 6 . | | -27 | highest stake allowed. |
| 15 . | | -30 | 9 . | | -18 | |
| 16 . | | -14 | 12 . | | -6 | |
| 15 | . | -29 | 7 | . | -13 | As explained before. |
| 16 . | | -13 | 10 . | | -3 | |
| 14 . | | +1 | 4 . | | +1 | As explained before. |
+------+-----+------+------+-----+------+------------------------------+

Had the player lost 60 units, he would have re-started the system and played 4, 8, 12, &c.; and if this play showed a net loss of 100 units, 5, 10, 15, &c., {461} would have been staked, and continued with until either the net loss was 150, or the net gain 1 unit, in which case the player would begin all over again with a single unit.

Another style of play is to bet on the prospect of the colour, or even chances, running in a particular way. Some people play for an intermittence of colour, consequently always stake on the opposite colour to that which turned up last. Others play for the run, and so always stake on the colour that last appeared. A very popular wager is to stake on the "_Avant dernier_," or on the colour that turned up the last time but one. By this means there is only one combination of events by which the player loses, and this is if the colours go two of one kind, followed by two of the other; but the weak point about it is that the player may miss his first stake and his last one, although the series goes in his favour. Yet another common method of staking is to play "the card"--that is, to play in expectation of previous events repeating themselves. Thus if the previous throws have given three Blacks, followed by three Reds, the expectation is if three Blacks immediately occur, that three Reds will also occur.[110] Such theories, of course, have absolutely no scientific basis, and, in the opinion of the writer, are only vexatious and a cause of trouble to the player, who should invariably stake on the chance that is most convenient to where he is sitting. He has an equal chance of winning, and by this means will save himself the trouble of reaching across the table, both to place his stake and to retrieve his winnings.

{462}

There may be, however, some reason in playing for a run on one colour or chance, but _not staking_ until after this colour or chance has appeared. By this means the player, if he plays flat stakes, is square on all runs of two, wins one on all runs of three, two on all runs of four, and so on. He loses one unit on every _intermittence_, but against this he loses nothing at all on all runs of the opposite colour or chance.

Had this method of staking been followed in the example given on p. 460, it will be seen that the player would have won 2 units on Red and 4 units on Black, and the highest stake necessary on any coup would have been 3 units; and had it been adopted in the example given on p. 457, only 70 units would have been lost on the Red side, and the highest stake risked 16; while on the Black, 41 units would have been won, with 9 as the highest stake.

It is advisable, when playing a system, to play on both sides of the table at once. The calculations for both Red and Black are kept, and the differences staked on the Red or Black as the case may be. The writer has actually seen a player stake the full requisite amount demanded by his system on both Red and Black _at the same time_. This of course gives the same net result as staking the difference on one colour, provided zero does not turn up. If it does, however, the player loses one-half of two large stakes in the one case, instead of only one-half of a small stake in the other case.

The advantage of playing a system on both sides of the table at the same time is that double as much can be won with the same capital that is required for playing on one side only. Indeed, slightly less capital is required, for obviously the player must {463} be winning something on one side to go against his loss on the other. The objection, of course, to this dual system of play is, that there is a double chance of striking an adverse run.

While on the subject of where to stake one's money, the reader, if a novice at Monte Carlo, is recommended to hand the amount of his wager to one of the croupiers to place on the table for him. This will ensure both the money being placed exactly as the punter desires, and the receipt of any winnings, without disputes on the part of other players. Unless one's French accent is above reproach, it is advisable to talk English to the croupiers. The writer, wishing to stake on Nos. 3, 12, and 15 on one occasion, handed the _chef-de-partie_ three 5-franc pieces, saying, "_Sur le 3, 12, 15, s'il vous plait._" After a short conversation on the subject the _chef_ said in perfect English, "If monsieur will please speak English, I will see that his money is correctly staked."

* * * * *

{464}

TRENTE ET QUARANTE.

BY CAPTAIN BROWNING.

TRENTE ET QUARANTE is played with six packs of cards on a table marked out as in the illustration (Fig. 3); this represents one-half of the table, the other half being marked out in an exactly similar manner. There are but four chances--_Rouge_, _Noir_, _Couleur_, and _Inverse_, which are played on in the following manner. The six packs of cards, having been well shuffled, are cut, and so many cards dealt out face upwards in a row until the sum of the pips (Aces, Kings, Queens, Knaves, and tens counting ten each, and the Ace one) _exceeds_ 30 in number. Then a second row is dealt out in a similar manner, below the first one, until the number of the pips in this second row also _exceeds_ 30. The top row is called "Black," the second or underneath row "Red," and the Red or Blacks win according to which row contains the fewer number of pips--_e.g._ whichever row of cards adds up nearest to 30.

The number to which each row adds up is called "the point," and it will be plain that the best point possible is 31, and the worst point possible 40. It is customary, when calling out the "point" of Black and Red to drop the "thirty" and say simply 2 and 6, which would mean that the point of Black amounts to 32, and the point of Red 36, in which case the Black or top row would win. The Black "point" is always called out first.

{465} The other chance, the _Couleur_ and _Inverse_, is decided by the colour of the _first_ card turned up. If the colour of this card corresponds with the colour of the winning row, then _Couleur_ wins; if it is of the opposite colour, then _Inverse_ wins. Thus suppose the top or Black row of cards amounts to 35, and the _first_ card in this row is a _Black_ card, and the Red row amounts to 36, then Black and _Couleur_ would win; had the first card in the Black row been a Red card, then _Inverse_ would have won, being of the opposite colour to the winning row (Black).

The players wishing to back any particular chance place their stakes on that portion of the table reserved for Black, Red, _Couleur_, or _Inverse_, as shown in the illustration (Fig. 3). There are two _chefs-de-parties_ employed to supervise the game, and four croupiers to receive the losing stakes and pay the winning ones, one of the croupiers also being the _tailleur_, or dealer of the cards. The _tailleur_ calls the game by saying, "_Messieurs, faites vos jeux_," when the players stake on the different chances. He then says, "_Les jeux sont fait. Rien ne va plus_," after which no further stakes may be made. He then deals out the cards, and when both rows are complete he calls the result thus, "_Deux, six, Rouge perds et Couleur gagne_," or "_Rouge perds et Couleur_," as the case may be, meaning that the point of Black is 32 and that of Red 36, so that Black and the colour win; or Black wins and the colour loses. It should be noted that the "_tailleur_" never mentions the words "Black" or "_Inverse_," but always says that _Red_ wins or _Red_ loses, and that _the colour_ wins or _the colour_ loses. On the conclusion {466} of each coup both rows of cards are swept into a small basket called the "_talon_," which is let into the centre of the table, and the game begins again. When the six packs of cards are exhausted, the "_tailleur_" says, "_Monsieur, les cartes passent_," when all the cards are collected out of the _talons_, re-shuffled and cut, and a fresh deal is started.

All four chances--Red, Black, _Couleur_, and _Inverse_--are of course even chances, and are paid as such by the Bank; but should the total (or point) of both rows of cards be exactly 31 each, the same procedure occurs as upon the appearance of the zero at Roulette--that is to say, the stakes are put _en prison_; then another deal is made, and those stakes which are on the winning chances are allowed to be withdrawn by the players. Or, as at Roulette, the stakes, at the players' option, may be halved with the Banker in the first instance.

Saving 31, all other identical points made by the Red and Black cause that deal to be null and void, the player being at liberty to remove his stake or otherwise, as he chooses. The condition of affairs (both rows coming to 31 each) which corresponds to the Roulette zero is called a "_Refait_," and is announced, as are all other identities of the points, by the word "_apres_." Thus suppose the Black row counts up to 38, and the Red row to the same figure, the _tailleur_ announces "_Huit, huit apres_." If it happens to be a _Refait_, he says, "_Un, un apres_," and the stakes are put into prison.

The _Refait_ is _said_ to occur once in 38 deals on the average; and if this were true, the Bank would have a slightly less advantage at Trente et Quarante than it has at Roulette. To arrive at the mathematical odds in favour of the Bank would involve an exceedingly {467} complicated calculation, and it is doubtful if they have ever been exactly computed. At a glance it would seem that the odds against both rows being 31 each is 81 to 1; there being 10 possible points for each row, the chances against any named point appearing would seem to be 9 to 1, in which case, of course, the chances against _both_ points being identical would be 9 x 9, or 81 to 1. But as the point of 31 can be formed in 10 ways--for the last card may be of any value, while the point of 32 can only be formed in 9 ways--for now the last card cannot be an ace; and to form a point of 33 the last card can be neither an ace nor a deuce, and so on with every point up to 40, which can only be formed in one way--viz. when the last card is a 10--it is obvious that 31 is the easiest possible point to arrive at, and the exact chances against its formation have, as far as the writer's information goes, never been calculated.[111]

In actual play, however, the punter may insure against the _Refait_ by paying a premium of 1 per cent. on his stake (at a minimum cost of five francs); thus it is safe to assume that for all practical purposes the percentage in favour of the Bank is exactly 2 percent.[112] Thus it would seem that once in 38 is an underestimate of the appearance of a _Refait_.

The maximum and minimum stakes allowed at Trente et Quarante are 12,000 francs and 20 francs respectively. Much heavier amounts are to be seen at stake at this game than at Roulette. This probably arises from two facts: because the games are generally {468} carried out in a quieter manner and the coups are more quickly played than is the case at Roulette, and because there is unquestionably a prevailing idea amongst the gamblers at Monte Carlo that the Bank's advantage is not so great at Trente et Quarante as it is at Roulette. The latter consideration is probably wrong; and, as far as the writer's experience goes, it is a very paying business to insure the stake at Trente et Quarante. If this really is so, it follows that the percentage in favour of the Bank is over 2 per cent., or something like 1 per cent. _more_ than it is at Roulette.

Any system that is applicable to the even chances at the Roulette table can of course be played at Trente et Quarante; but for some reason or other it is unusual to see any system properly worked at this game, possibly because too large a capital would be required.

The almost universal method of play is to follow the "_tableau_"--that is, to follow the pattern of the card on which the game is marked. If there have been two Reds followed by two Blacks, ninety-nine people out of a hundred will stake on Red, in the expectation of two Reds now appearing, while if there is a run of one colour, thousands of francs will be seen on that colour, and not a single 20-franc piece on the other. Sometimes the colours do run in the most inexplicable manner at Trente et Quarante. The writer has played at a table where there were 17 consecutive Blacks, then 1 Red, to be followed by 16 consecutive Blacks. When such runs occur, the Banks of course lose heavily, and are constantly broken. To break the Bank in the true sense of the word is of course an impossibility. When a Bank gets into low water the _chef-de-partie_ {469} sends for some more money, which is "_Ajouter a la banque_," and to this extent only is it possible to "break the Bank at Monte Carlo."

The game of Trente et Quarante is sometimes called "Rouge et Noir."

The method of play on the even chances that will now be explained is based on the three following assumptions:--

First. That every system at present played is successful only for a certain time, when an adverse run, long enough to defeat the progression adopted, is almost certain to occur, whereby the Bank reaps a rich harvest.

Secondly. That only on rare occasions does the system show the desired profit, without the player having been at some period of the game a very heavy loser.

Thirdly. That the failure of systems is not due to zero, but to the Bank's maximum.

These conditions are _assumed_, though in the first two cases they undoubtedly are realities, and within the experience of every system player. The third one may be true or not; it is not vastly important.[113]

Now as regards maxim No. 1, it may be taken for granted that for all practical purposes the system player makes his "_grand coup_"[114] on not more than {470} (say) twenty occasions, and on the twenty-first he meets such an adverse run that he loses his entire profits plus his entire capital; or say, for argument, he had already spent his profits and so loses only his entire capital. The proportion of the coup played for to the capital employed is generally some 2-1/2 per cent.; consequently after twenty good days' play, and one bad one, a system player is a loser of 50 per cent. of his money. (This is a very low estimate.)

Now supposing a player had played stake for stake on the opposite chance to that played on by the system player, it is obvious that he would have lost on twenty days, and won on the twenty-first sufficient to recoup all his previous losses, with 50 per cent. profit.

The mathematician will say "No" to this--"the Bank will have reaped its zero percentage from each spin of the Wheel during the progress of the play." But why? A, who is playing the system, stakes 10 louis on Red; B (who is playing against him) stakes 10 louis on Black, and zero crops up. They are both put in prison, and A comes out safely, so B is now 10 louis worse off than A. But in a short time A and B again both stake 10 louis, and zero appears. But this time B comes out safely, in which case A must write this down as a losing coup, and his next stake will be say, for example, 15. To meet this B has only to add 5 louis to the 10 he has just retrieved out of prison--so his profit and loss account due to zero is exactly square, as far as it affects his transactions with A. And surely during the course of a game A and B will both get out of prison the same number of times. (And A does not fear zero--he only fears reaching the maximum--consequently B {471} does fear for zero; he but awaits the time when his stake gets to the maximum.)

Is it not desirable to be B? He requires no capital--or very little--and yet is in a position to win all that A is eventually going to lose--as he most certainly _must_ lose. To play on this method is exceedingly simple. All that has to be done is to take _any_ system, and play it in reverse order to what it is designed to be played in. The effect of this is, in a word, to compel the Bank to play this system in its correct order against the punter. The writer has always employed a _Labouchere_ to play on this method, and it is the simplest one by which to explain the procedure.

A reference to p. 456 will show that the _Labouchere_ system, is played by writing down so many figures, so that their sum amounts to the _grand coup_--or stake being played for--and that it is usual to write down the figures 1, 2, 3, 4; so that the _grand coup_ is 10 units. To play this system in the usual manner it is generally assumed that a capital of 400 or 500 units is required. By reversing matters in play the first important advantage gained to the player is that he needs but a capital of 10 units, and his _grand coup_ becomes 400 or 500 units. Very well. The figures 1, 2, 3, 4 are written down, and the first stake is the sum of the extreme figures--5. This sum is lost; but now the 5 is not written down after the 4, but the _1 and the 4 are erased_. The next state is again 5 (2 + 3), and is again lost, the 2 and 3 are erased and the player retires. Suppose this second stake of 5 had been won, then instead of erasing the 2 and 3, the figure 5 would be written down on the paper, so the row would read =1=, 2, 3, =4=, 5, and the next stake would be (5 + 2) 7. Should this be lost the 5 and 2 are {472} erased, the next stake being 3. Suppose it is won, this figure is written down, and the row now reads =1=, =2=, 3, =4=, =5=, 3, and the next stake is 3 + 3 (6), and so on. But the moment all figures are erased, the player will have lost 10 units and must retire. This he will have to do a great many times, but finally such a run as the following will occur. The Red is staked on throughout--the dot indicating which colour wins.

Figures. Stake. R. B. + or -
=1= 1 + 4 5 . -5
=2= 2 + 3 5 . 0
=3= 2 + 5 7 . +7
=4= 2 + 7 9 . +16
=5= 2 + 9 11 . +27
=7= 2 + 11 13 . +14
=9= 3 + 9 12 . +2
=11= 5 + 7 12 . +14
=12= 5 + 12 17 . +31
=17= 5 + 17 22 . +53
=22= 5 + 22 27 . +80
=27= 5 + 27 32 . +48
7 + 22 29 . +19
=29= 12 + 17 29 . +48
=41= 12 + 29 41 . +89
12 + 41 53 . +36
=46= 17 + 29 46 . +82
17 + 46 63 . +19
=29= 29 29 . +48
58 29 + 29 58 . +106
=87= 29 + 58 87 . +193
29 + 87 116 . +77
=87= 29 + 58 87 . +164
29 + 87 116 . +48
58 58 58 . +106
116 58 + 58 116 . +222
174 58 + 116 174 . +396
232 58 + 174 232 . +628
290 58 + 232 290 . +918

This shows a run of 29 coups, of which the player wins 20 and loses 9. {473}

He is 918 units to the good, and his next stake would be 348![115]

Assuming a player had been working a _Labouchere_ on this run in the usual manner, on Black with a capital of 500 units, he would have had to retire after the 27th coup through lack of capital; and assuming him to have been playing with a 20-franc unit, he would have had to retire from Roulette on the 28th coup, and from Trente et Quarante after a few more coups if the bad sequence continued, no matter how large his capital had been.

It has been stated that the Bank beats the system player only on account of its limit. This is not quite true; it has also one more great advantage over the player, and this is the fact of its being a machine, while the punter is human; and although a player will stake his all to retrieve his previous losses, he will not--nature will not allow him to--risk his winnings to win still more.

This is a psychological fact that cannot be explained. It must be to the knowledge of most people who have visited Monte Carlo, that a player will stake as much as 500 francs to retrieve a loss of a single 5-franc piece. Yet the same player, having turned a 5-franc piece into as little as 50 francs, will refuse to adventure another stake, and retire from the gaming-table. When the player is having his bad run, the Bank cannot help playing their winnings to the maximum stake--they _must_ do so; but the player on his good run is not compelled to play up his winnings, and really cannot be expected to do so. Theoretically {474} he should, and I firmly believe there is a lot of money awaiting the player who has the patience to wait for such a run--which must come to him, equally as it must and does, we know, come to the Bank--and then play on and on until he is prohibited by the Bank from staking any higher. To play a system upside-down, or in reverse order, requires great patience and equanimity, until the favourable run occurs, when indomitable pluck and perseverance are the necessary qualifications.

The writer feels bound to take the reader into his confidence so far as to acknowledge that he himself has never had such pluck, but has always retired on winning between 200 and 300 units. But he has always watched the future run of the table, and on no less than five occasions would have reached the maximum stake and won over 1000 units. He has, however, always had the patience, and lost his _petit coup_ time after time with perfect equanimity, and only wishes he had had the other qualifications as well.

Referring for one moment to the assumed fact No. 2 on which this method is based--that a player more often than not is in deep water before bringing off his _grand coup_; which he must be, owing to the losses being so disproportionate in magnitude to the gains--it might be a good plan to discover what the average highest loss of a system player is before the system shows a profit, and then to play the same system in reverse or upside-down order, making this figure the _grand coup_. Playing in this manner, a visitor will have a cheap and enjoyable visit to Monte Carlo, and may be assured of one of the most exciting little periods of his career when this favourable run of luck does come his way. {475}

One final word of advice to all system players. Play on the chance that is most convenient to your seat at the table. It is as likely to win as any other. Never get flurried with your system or calculations. It is not at all necessary to stake on every coup. You are just as likely to win if you postpone staking until the day after to-morrow, as if you stake on the very next spin of the Wheel--the Rooms are open for twelve hours per diem, which should allow ample time for the number of coups you wish to play.

There may or not be such a thing as "luck." There can, however, be no harm in giving its existence the benefit of the doubt. If on some particular occasions you find you cannot do right, _assume_ you are out of luck, and stop playing. Do not consider either that you owe a grudge to the Bank because you have lost, or that it is absolutely necessary to retrieve your fortune then and there! Postpone playing until the following day, or week, or year, when you may be in _good luck_, and can easily recoup yourself.

Always bear the clever gambler's great maxim well in mind: "Cut your losses--play up your gains!"

The writer's only object has been to try and explain how the games of chance are played at Monte Carlo, and to point out that the player is at a disadvantage on each occasion that he stakes, though that disadvantage may be increased or reduced by bad or good staking. It now remains for the reader to decide whether the pleasure he derives from gambling is likely to recompense him for his probable losses.

Printed by BALLANTYNE, HANSON & CO.
Edinburgh & London

* * * * *

NOTES

[1] This is the old-fashioned rule, but at the present day the Whist rule of "lowest card deals" is frequently followed.

[2] See note on last page.

[3] For the accepted Laws of All-Fours, see _The Book of Card and Table Games_ (Routledge).

[4] Pronounced _Back[)a]rah_.

[5] The number is not absolute, sometimes four packs, sometimes two only, being used; but three is the more usual number.

[6] For the Laws of _Baccarat Banque_, and some suggestions for play, see _The Book of Card and Table Games_.

[7] Some players do not score _brisques_ till the close of the hand. The better rule, however, it to score them when the trick is won.

[8] In some circles, when the Whist tricks are reached, the ten reverts to its Whist rank, _i.e._ below the knave, but the practice is not recommended.

[9] _Carte blanche_ is scored at the outset of the game, and before the player has drawn a card. He must prove his title by exhibiting his nine cards, one after another (as rapidly as he pleases), face upwards on the table. Should the first card he draws not be an honour, he may show the card, and again score _carte blanche_, and so on, as often as this may happen; but _carte blanche_ cannot be scored after the player has once held a court card.

[10] The first marriage scored is necessarily in trumps.

[11] It will be observed that this rule is directly contrary to that prevailing at ordinary Bezique.

[12] Roughly, the value of all the brisques in the four packs. There are actually 32, which at ten each would be 320; but as the odd 20 are not reckoned, this reduces the value to 300.

[13] As a matter of fact, this arrangement is no guarantee whatever against pre-arranged fraud. For the methods employed by card-sharpers at this game, see _Les Filouteries du Jeu_ (Cavaille). Tit. "Les Petits Paquets."

[14] Court cards, though they all count as of the same value--_i.e._ "ten"--retain their distinctive rank for pairing purposes. Thus a knave can only be paired with a knave, and so on.

[15] A single fifteen is spoken of as fifteen two, two fifteens as fifteen four, three as fifteen six, and so on. Four (fifteen eight) is the largest number of fifteens that can be made with four cards.

[16] If the knave and start be of different suits, the score is twenty-eight. With four fives in the crib, and the knave turned up, the value of the show will be twenty-eight only, but the dealer will already have scored "two for his heels," so that the total value is thirty.

[17] The score is made up as follows. Each of the sixes combines with each nine to make a fifteen, giving fifteen four. Again, each of the threes combines with the two sixes, bringing the score to fifteen ten. The pair and pair-royal make it eighteen.

[18] If the three tenth cards make neither pair nor sequence, the score will be fourteen only.

[19] In the case supposed, it would be very unwise for A to pair the eight, as, in the event of B's holding a second eight, he would make a "pair-royal" and "go" simultaneously.

[20] There is no authoritative code of Cribbage Laws, and there is considerable divergence of opinion on sundry minor points. For the rules generally accepted, the reader may be referred to the _Book of Card and Table Games_ (Routledge), tit. "Cribbage."

[21] De la Rue & Co.

[22] The elder hand may "propose," _i.e._ ask for cards, as often as he pleases. If the dealer is not content with his own hand, he will give cards, but after the first proposal, it is entirely at his own option whether or not to do so.

[23] For some further rules, defining the position and obligations of bystanders betting on the game, see the work of "Cavendish" referred to at p. 53.

[24] A still higher trump is sometimes by agreement introduced in the shape of a blank card, backed like the rest of the pack which in this case consists of thirty-three cards. This is known as the "Joker," or "Best Bower," and takes precedence even of Right Bower. If the "Joker" chance to be turned up, the card next in order decides the trump suit.

[25] Under the more modern practice the player having the later call _can_ play alone in place of his partner. Only a very strong hand, however, would justify his doing so.

[26] There is no English Code of Laws for Euchre. The accepted American Code was compiled in 1888 for the Somerset Club, Boston, Massachusetts, by Messrs. H. C. Leeds and James Dwight. It will be found reprinted at length, by their permission, in the _Book of Card and Table Games_.

[27] This is usually done by dealing a preliminary round, face upwards, the first knave turned up entitling the holder to the deal.

[28] As, for instance, where the player holds the seven and nine of trumps, the eight having been turned up; the seven and nine are then of equal value.

[29] Sometimes the preference is given to the elder hand, irrespective of the value of the cards.

[30] The words between brackets apply of course to three-card loo. Sometimes the dealer is allowed, after dealing one card to each player, to deal three together for a miss, but the practice is irregular.

At five-card Loo the _Ecarte_ method of dealing (first by threes, and then by twos, or _vice versa_) is sometimes adopted.

[31] For an instructive series of illustrative hands at Napoleon, see the _Book of Card and Table Games_.

[32] A having made seven out of twelve.

[33] See in particular the excellent treatise on the game by "Cavendish," published by Messrs. De La Rue & Co.

[34] For the authorised Laws of the Game, in its modern form, see _The Book of Card and Table Games_, or the treatise of "Cavendish" before mentioned.

[35] As the game is sometimes played, the dealer, and not the Age, puts up the _ante_, but the contrary is the more usual practice.

[36] This being a compulsory stake on an unknown hand, it is prudent to make it as small as possible.

[37] The Age, as a rule, goes in, even with poor cards; if he passes, he is bound to lose the half stake already put up, and it is, therefore, generally worth his while to risk the other half.

[38] Should B have already thrown up his cards, the privilege does _not_ pass to C. There is a maxim on this point, "The Age never passes."

[39] Some players on a second round only allow the jack-pot to be opened by a pair of queens, or better; on a third, only by a pair of kings, or better; and on a fourth, only by a pair of aces, or better; but the practice is not recommended.

No player, even though holding the needful cards, is bound to open the jack-pot unless he pleases.

[40] Strictly speaking, each dealer in rotation should himself dress the board, but it will be found more convenient to depute some one player to do so throughout the game.

[41] By some players the dealer is allowed the privilege of looking at the extra cards (sometimes, but incorrectly, themselves spoken of as "the stops"), and to act as a kind of referee as to whether a given card is a stop or otherwise, but the practice is not recommended.

[42] The Misere is now introduced into Napoleon. See p. 96.

[43] For more minute information, and for a number of illustrative hands, see _The Book of Card and Table Games_.

[44] The right to deal is usually decided by a preliminary deal of faced cards, the first ace, or first knave, as may be agreed, having the preference.

In some circles, after the cards are cut, the dealer is allowed to look at the bottom card, and if such card prove to be an ace or tenth card, he also looks at the top card. If the two form a "natural," he is entitled to receive double the _minimum_ stake all round.

This privilege is known as the _brulet_, from the fact that it is dependent on the nature of the bottom card, which is always, in the French phrase, _brule_ (literally, "burnt") _i.e._ thrown aside when reached in the course of the deal, and not dealt to any player.

The _brulet_ has never been recognised as an essential part of the game, and is now generally abandoned.

[45] Some players risk the maximum stake on a seven, but we question the expediency of doing so.

[46] This amount is the same as is paid for an ordinary Vingt-Un, _i.e._ one made with more than two cards. Sometimes, by agreement, a "natural" receives double the amount of an ordinary.

[47] Many players habitually stand at fifteen, and if the dealer is a reckless player, with a tendency to overdraw, it may be good policy to stand upon an even smaller figure. "Cavendish" is in favour of standing, as a rule, on fifteen.

[48] Pronounced like _pony_.

[49] _Example._ A three, two sixes, and a knave are drawn. The two sixes draw again, and the lower plays with the three. Suppose, at the second draw, the two sixes draw a king and a queen, the queen plays with the three.

If at the second draw, a lower card than the three is drawn, the three still retains its privileges as original low, and has the deal and choice of cards and seats.

[50] _Example._ Three aces and a two are drawn. The three aces draw again. The two is the original high, and plays with the highest of the next draw.

Suppose, at the second draw, two more twos and a king are drawn. The king plays with the original two, and the other pair of twos draw again for deal.

Suppose, instead, the second draw to consist of an ace and two knaves. The two knaves draw again, and the higher plays with the two.

[51] _Vide_ Law 26.

[52] _Vide_ Law 29.

[53] After the two packets have been re-united, Law 30 comes into operation.

[54] _Vide_ also Laws 36 and 41.

[55] _Vide_ also Law 28.

[56] The pack being perfect. _Vide_ Law 41.

[57] Except as provided in Law 36.

[58] It is not usual to call the trump card if left on the table.

[59] _Vide_ Law 75.

[60] _E.g._, If a single is scored by mistake for a double or treble, or _vice versa_.

[61] _Vide_ also Law 40.

[62] _Vide_ Law 81.

[63] The more complicated forms of the so-called "American" leads are not set out, as they never gained general acceptance.

[64] This penalty is not affected by a double.

[65] Pronounced _tray_, _kater_, _sank_, and _size_, respectively.

[66] This applies more particularly towards the close of the game. The leaving of a blot at the outset, when five out of six of the points in the adversary's table are still open, is a comparatively unimportant matter.

[67] This leaves a blot on the deuce point in your outer table, but this is a trifling disadvantage as compared with the gain of at once securing four points side by side. There are only three throws, six ace, cinque deuce, and quatre trois, that will enable the adversary to hit the blot; and your next throw will in all probability enable you to place it beyond the reach of danger, either by playing another man on the same point, or by transferring the solitary man to one of the points already made.

[68] For further information as to the game and its chances, see the article on Backgammon in _The Book of Card and Table Games_ (Routledge), of which the present paper is an abridgment.

[69] For a description of other forms of the game, see _The Book of Card and Table Games_ (Routledge).

[70] See p. 296.

[71] Throughout these rules, "coloured balls" mean the six balls (not Red) specified in Rule 2.

[72] _Vide_ page 290, Definition 4.

[73] For fuller information on the subject of the game, see Mr. L. Hoffer's excellent treatise on Chess in _The Book of Card and Table Games_ (reprinted separately in the Oval Series, Routledge. 1''s.''), of which this section is a much condensed abridgment.

[74] For the meaning of these letters and figures, see Chess Notation (p. 343).

[75] This is possible in case of a check with Queen, Rook, or Bishop, but not in case of check with a Knight or pawns.

[76] From the Italian _Gambetto_, "a trip up."

[77] With two Bishops checkmate can be forced, whilst with two Knights only checkmate cannot be given against the best defence.

[78] Intending to establish a centre at once.

[79] Considered the best reply. Black develops a piece and attacks a pawn.

[80] Or 8. ... Kt takes P; 9. Kt takes Kt, P to Q 4; 10. B to Q 3; P takes Kt; 11. B takes P, Kt to K 2, &c.

[81] If 11., Kt takes R, Black would proceed with 11. ... Q to K 2; 12. Kt to B 7, B takes P; ch.; 13. R takes B, P takes R; ch.; 14. K takes P, Kt to Kt 5; ch.; 15. K to Kt 3, Q to B 3; 16. Q to B 3, Q to Kt 2, &c., with a powerful attack.

[82] A safe defence, though troublesome for a time.

[83] The best move, White threatening with 11. P to Q 5 to win a piece.

[84] 12. B to K 2 is a sounder move.

[85] The best move. 12. ... P to Q R 3 may also be played; but not 12. ... P to Q R 4, because it weakens the pawns on the Queen's side for the End game.

[86] Because Black threatens 15. ... Kt to K 4, and after 16. ... B or Kt takes Kt; 17. P takes Kt, Q to Kt 4: ch., winning the K P.

[87] This move is inferior to 9. P to K 5.

[88] The best move. 11. ... Kt to B 3 would be inferior.

[89] Black gives up a pawn for a temporary counter-attack: It is a safer defence than 5. ... Kt takes P.

[90] The only right square for the Bishop, because it secures a retreat for the Knight on both sides, as will be seen by the sequel.

[91] If 7. ... P takes B, White gets the piece back with 8. Kt takes Kt, P takes Kt; 9. R to K sq., &c.

[92] Here again, if 8. P takes B, White replies 9. Q to Q 5, &c.

[93] As before, if 9. ... P takes B; 10. Q to Q 5 follows.

[94] The original move upon which the opening was based. But it is unsound, as the two specimens given sufficiently prove. The alternative continuation is 3. ... B to K 2; 4. P to Q 3, followed by 5. Kt to B 3, &c.

[95] If 6. ... R takes Kt, then 7. Q to R 5: ch., K to Q 2; 8. B takes Kt, R takes B; 9. Q takes P: ch., and wherever the King moves the Queen mates.

[96] If instead of the text move 11. ... P to B 5, White wins with 12. R to K sq.

[97] Threatening 9. Q to Kt 5; ch., K to Q 3; 10. B to B 4: ch. and 11. Q to K 5: ch., &c.

[98] 10. ... P to B 4 would be immediately fatal.

[99] If 12. ... K takes B, then 13. Q takes K P: ch., K to Kt 4; 14. Kt to B 3. ch., K to Kt 3; 15. Q to Q 4; ch., &c.

[100] If 15. ... Q takes Q: then 16. R to R 4: ch., K moves; 17. Kt mates either at Kt 3 or at Kt sq. accordingly.

[101] Attacking the Rook.

[102] 11. P to K Kt 3 would be bad, because of 11. ... P takes P; 12. Q takes P, R to B sq.: ch.; 13. K to Kt 2, Q to R 4, threatening 14. ... B to R 5, winning.

[103] Not 12. P takes P, because in such case Black replies 12. ... P to Kt 6, and wins.

[104] If 7. Q to B 3, Black replies 7. ... P to Kt 7: ch.; 8. K takes P, P to K B 3; 9. B takes Kt, R takes B; 10. Q takes P, B to K 2; 11. Q to K B 3, R to B sq., with the better position.

[105] Meaning _Flank_.

[106] The work of which the present article is an abridgment. The Openings here given will be found treated in this book at much greater length, with others scarcely less valuable, and a fund of general Chess information.

[107] In England it was formerly the custom to play on the white squares, but the Scottish practice of using the black squares is now generally adopted. So far as the course of play is concerned, the one plan is as good as the other; and in all treatises on the game the men are, for typographical reasons, shown on the _white_ squares. This involves a corresponding alteration of the position of the board, which is shown with a _white_ bottom square on the left hand.

[108] A player may be huffed for not taking the full number of men he should have taken by the play adopted. Thus if he takes one man only, where by the same play, duly continued, he could have taken two, he is liable to the huff. If, however, he has the choice of two moves, by one of which he would take a larger number of men taken than by the other, he is under no obligation to adopt that move.

[109] See p. 469, footnote.

[110] This is a more common method of play at "Trente et Quarante" (see p. 468).

[111] A German mathematician is said to have calculated the percentage in favour of the Banks to be 1.28 per cent.

[112] It must be remembered that as the player is at liberty to withdraw half his stake when there is a _Refait_, he is really paying a premium of 1 per cent. to insure only _half_ his stake.

[113] If there were no limit every one could win at Monte Carlo, by the simple method of doubling up after each loss. Hence sans maximum, zero does not prevent the Bank from losing.

[114] Most system players try to win a percentage of their capital per diem. Having done so, they retire from the table. By "_grand coup_" is meant this amount of daily winnings. There is no reason why a player should not play his system _ad infinitum_. He, however, instinctively knows the grave risk he is running by continuing his game, and is generally very pleased to retire after having made a certain daily profit.

[115] In the series shown on p. 457, had a player been fortunate enough to have played a "_Labouchere_ reversed" on Black, he would have won 890 units.

* * * * *

Transcriber's note:

Corrections made to printed original:

P. 41 (Score for Three tenth cards in sequence and two
fives):--"17", printed as "11" in original.

P. 127 (in "a flush or sequence lacking one card"):--"card",
printed as "care" in original.

P. 354 (in "The Kt here attacks Black's K P"):--"Black",
printed as "White" in original.

P. 371 in "he cannot play 20. ... R takes R" the first R
is missing in printed copy and has been restored by
considering the position.

Additional material for Project Gutenberg Edition

Forsyth-Edwards Notation for Chess and Draughts figures

CHESS

Fig. 5. Drawing by perpetual check.
1k6/2p5/QpB5/1P6/8/P2b4/1P6/K1n5

Fig. 6. Illustration of stalemate.
1k6/1P6/1K6/8/8/8/8/8

Fig. 7. Giuoco Piano. Position after Black's 15th move.
r2nk2r/pppbqp2/1b3npp/4p3/4P3/2P1BNN1/PPB1QPPP/R4RK1

Fig. 8. Giuoco Piano. Position after White's 28th move.
3r4/1pp2p1k/1pn2npp/4pN2/2q1P3/P1P1Q1NP/1P3PP1/5RK1

Fig. 9. Evans Gambit. Position after White's 19th Move.
1rbq1rk1/p1b3pp/3p1pn1/np1P1N2/2p1P3/5N2/PB1QBPPP/2R2R1K

Fig. 10. Two Knights' Defence. Position after Black's 9th move.
r1bq1b1r/pp2n1pp/2p1k3/3np3/2BP4/2N2Q2/PPP2PPP/R1B1K2R

Fig. 11. Scotch Gambit. Position after White's 11th Move.
r1bqr3/ppp2k1p/2n3p1/2QP4/3p4/8/PPP2PPP/RNBQ3R

Fig. 12. Muzio Gambit. Position after Black's 5th Move.
rnbqkbnr/pppp1p1p/8/8/2B1Pp2/5p2/PPPP2PP/RNBQ1RK1

Fig. 13. White pawn advantage. Black to Move and Draw.
r7/8/q7/8/8/1P5k/1R6/6QK

Fig. 14. King and pawn against King. Black to Move and Draw.
8/8/8/8/8/1P5k/8/6K1

Fig. 15. King and pawn against King. White to Move and Win.
8/8/8/8/8/1k6/6P1/2K5

Fig. 16. King and Queen against King. White to Move and Win.
8/8/4k3/8/8/8/8/KQ6

Fig. 17. Two Rooks and King against King. White to Move and Win.
8/8/8/3k4/8/8/8/KRR5

Fig. 18. King and Rook against King. White to Move and Win.
8/8/8/4k3/8/2K5/8/1R6

Fig. 19. King and two Bishops against King. White to Move and Win.
8/8/8/8/5kBB/8/8/K7

Fig. 20. King, Bishop, and Knight against King. White to Move and
Win. 8/8/8/K1k4B/8/8/8/N7

DRAUGHTS

Fig. 3. Two Kings to One. White to Move and Win.
WK11,K22:BK5

Fig. 4. Three Kings to Two. White to Move and Win.
WK11,K15,K18:BK5,K28

Fig. 5. First Position. Black to Move and Win.
W30,K8:B22,27