## Wj On The Zarrow April 1970

Prepare for the Shuffle by undercutting more than a third but less than half of the deck to the right. The smaller the portion you take, the smaller the block you will have to screen later. Thus, unless the application demands it, slighdy less than half is preferable. (See notes, at the end, on the question of cutting the top or the bottom to the right.)

Bring the packets together so that they form an inner angle of 160-165 degrees. An angle smaller than 160 degrees may be suspect, but significantly greater angles make the unmesh more difficult.

Riffle the cards together lightly. The lighter the riffle the easier the unmesh. (See notes for further discussion of this point.) It is generally desirable to release a block of about five cards from the right-hand packet before "freely" interlacing the two portions. On hard surfaces, as many as ten cards are helpful. On hard or slippery surfaces, you may elect to forego this initial five-card block on the right in favor of an eight- to ten-card block on the left. In such cases, you would lay down the bed on the left, then the five on the right, after which the balance of the deck can be shuffled. It is better to allow the interlaced cards to fall in blocks of five to seven rather than to approach a card-for-card interweave, but it is not crucial. If the specific application for which you are using the Zarrow doesn't dictate otherwise, the last cards from the right hand should fall while three or four cards from the left-hand packet remain. You should know whether it is three or four, as you must shuffle under the same number in the second Shuffle. You can "slur" (spread over) the top few cards of the right-hand packet in a line to the left but in line with their packet. This slurring should not be pronounced, as it can be a tip-off to the Shuffle. The screen suggested in Vernon's description of the Zarrow is not necessary or, I believe, desirable.

Grasp the packets lightly with your second and third fingers positioned along the outer edges of their respective portions, the fourth fingers at the outer corners, first fingers curled onto the backs of the packets and the thumbs resting about a third of the way back from the adjacent ends (Figure 196).

Both hands now press down and move to the right and back, slightly slurring the packets (Figure 197); then, as if correcting for this, move the hands back to the left (Figure 198). This causes a set of yyy small steps to form along the sides of both packets, because of the bevels you've created, thus beginning to produce some edge break-up. These adjustments should be done without emphasis and as if by accident. They must seem incidental if they are noticed at all.

The side of the right thumb must make firm contact with the surface of the mat at a point under the near side of the right-hand packet. This position is critical. The thumb being under rather than at the side of the packet causes the right-hand packet to inscribe a wider arc than it otherwise would when the packets are rotated together, as they will be in a moment. This larger arc causes the unmeshing of the packets and continues the break-up of the edges. Moreover, positioning the right thumb under the packet prevents the right hand's cards from falling toward the table when they are unmeshed.

Both hands rotate on their rhumbs in an outward arc. It should require no more than about ten degrees of arc to unmesh the packets. As soon as the unmeshing is complete, stop the rotation but be sure the length of the right thumb supports its entire stock. Don't allow these cards to drop. Instead, the right thumb drives the lower cards of the packet up to join the top card, but the top card should not rise at all.

The surrounding fingers begin pushing toward each other at an acute angle of five degrees, without actually gripping their packets. At the same time, both hands move a bit to the left again, in a line parallel to the left-hand packet. This causes additional steps to form in the packets. The right-hand push, which happens fairly rapidly, ends when the corner of the right packet hits behind the left second fingertip. When the right packet encounters this fingertip, a binding is simulated that would occur in a genuine shuffle. The Shuffle would be vulnerable to scrutiny at this point if the fingers where not blocking the audience's view of the front edges of the deck. Be sure the fingers are angled to effectively screen the edges (Figure 199,where the corner of the right-hand packet is exposed for clarity). It isn't necessary for the fourth fingers, which are doing most of the screening of the corners at the moment, to touch the table surface to do their job effectively. They need only be lower then the top of the lowest packet.

Move both fourth fingers to the outer ends of the packets. The hands will assume a sharper angle to the table, nearer perpendicular, and will continue to mask the front edges of the packets to the right and left of center (Figure 200).

The fourth fingertips press against the table as you straighten your fingers until they are perpendicular to the tabletop. This drives the packets slightly inward. To the audience the deck will appear to be telescoping in as it would in a genuine shuffle. Eventually, the packets will be telescoped for all but a half inch on each side, which will be hidden by the third finger of each hand at the front. The fingers all around the packets now loosen their grip so that they form a frame around the deck without restricting its motion. The next movement is the most difficult to acquire. It's one of those "feel" things that abound in sleight-of-hand. It is also a most critical modification to the Shuffle. Both fourth fingers simultaneously, kick forward and toward each other. The inside edges of the fourth fingers lightly contact the ends of the extreme front corners of the packets. Figure 201 shows the fourth-finger positions before and after the kick action. Combined kicks from the fourth fingers and the way the packets, on both the left: and right, rebound off the other fingers, serve to produce what I call "packet shatter," making it appear as though an interweave had taken place. This touch finishes the telescoping of the

The hands now move away from the deck to take up positions alongside it for a normal side-squaring action. At the left rear corner you will see (and feel) a step. As you perform the squaring action, turn that step into a break. Your first Shuffle is complete.

Do a second Zarrow Shuffle, undercutting at the step and shuffling under the same number of cards you did in the first shuffle. Everything about this second Shuffle is identical to the first except, perhaps, the thickness of the packets. The deck is back in order; or, if you cut top to the right for the first Shuffle and bottom to the right for the second, they will be in order after cutting at the break. The break can be ignored after the second Shuffle if you've cut the bottom to the right for both Shuffles.

NOTES: For purposes of this discussion the Zarrow Shuffle is any Shuffle in which a table-riffled deck that has been meshed in the Shuffle is later unmeshed and inserted below one or more cover cards. I will not enter into the Zarrow-Shank debate.

When you shuffle, you should shuffle with a light touch. I believe this true for all shuffles, false or genuine; it makes the shuffles easier in their execution. In the Zarrow Shuffle, a heavy or tight shuffle will impede the unmeshing and encourage tip-off actions; for example, lifting or too big a swing. Conversely, too light a shuffle looks affected and can appear studied and suspicious. One should cultivate a happy medium. A general guideline is to always shuffle lightly enough to make a clean Push-Through easy, but no lighter.

I have for some years pondered the question of how many shuffles are required to convince a viewer beyond reasonable doubt that the deck has been mixed, and how many shuffles is too many. The numbers are, perforce, unprovable. Like the nuclear particle, they are also affected by the attempt at measurement (the Heisenberg Uncertainty Principle). The effect on the viewer is affected by a wide variety of psychological factors. Nevertheless, my empirical sense is that three is the most convincing and, therefore, the ideal number. The maximum is, in part, contingent on how quickly you shuffle, but in no case is it greater than six, with five being the general maximum.

My conviction that three shuffles are ideal has led me to create numerous three-shuffle combinations. I will confess that I most often use a Push-Through followed by a two-shuffle Zarrow, ending in a cut; or I might reverse the order of these shuffles. For purposes of this discussion, only the two Zarrows will be examined. I prefer cutting top to the right, so the top cards are seen to be buried in the first Shuffle. This choice also forces a cut after the second Shuffle, which is psychologically sound. If you prefer, or your application requires that you cut the bottom to the right, you can create the illusion of a top-to-the-right cut by using the following procedure:

Extend your right first finger about five-sixths of the way across the top of the deck. The left thumb forms a break at the point where you chose to divide the deck (Figure 202). The right thumb and fingers grip the lower block, and in one short hop-like