Generally credited to Edward Marlo, although this has been challenged by Karl Fulves. Marlo first published the effect and a method for it in M-U-M, Vol. 43 No. 4, Sep. 1953, p. 148. Later that year he published an expanded, three-phase routine in his book The Cardician, 1953, p. 111. Marlo went on to publish numerous methods and several variant effects.

Marlo has been quoted as writing that the Oil and Water routine was not new with him. However, the full quote from The Cardician, 1953, p. 112, reads: “Altho the Oil and Water routine is not new with us, having worked out several other versions years ago, this present version we feel is superior in directness of effects.” Here he seems to be saying that he has worked on the Oil and Water effect for years, and therefore it isn't new to him. Without the qualification given by the full sentence, the first phrase can seem to mean that he is acknowledging that Oil and Water was not his original effect.

Several color-separation tricks that preceded Marlo's “Oil and Water” have been cited to suggest that the plot was not original to him. The earliest is the centuries-old idea of separating the colors of a shuffled deck with one cut. The method employed a Stripper Deck. Reinhard Müller cites two early sources for this: William Hooper's Rational Recreations, Vol. 4, 1782, second edition, p. 264; and Ein Blick in Döbler’s und Bosco’s Zauberkabinett edited by L. Schellenberg, 1832, p. 193. Ottokar Fischer published what seems a description of this effect, rumored to have been done by J. N. Hofzinser with an unprepared deck; see “Die magische Separation” in Fischer's J. N. Hofzinser Kartenkünste, 1910, p. 217; also Magic Christian's J. N. Hofzinser: Non Plus Ultra, Vol. II, 2013, p. 390. There are three problems with this theoretical Hofzinser effect. First, it is uncertain if he actually did it. Second, the method is not known. And third, Fischer's description states that the performer sets the mixed deck on the table and makes two piles. It is not specified if the formation of these piles is done with a cut, by dealing or with some other procedure. With so many points of uncertainty, making a case for a relationship between this effect and Oil and Water is difficult.

Another trick proposed as a precursor to Oil and Water is Edward Victor's “Reversed Cards”, marketed in 1920 by Will Goldston. While no information on Victor's method seems to have survived, examining the ad for the trick in Goldston's Magazine of Magic, Vol. 7 No. 3, Feb. 1920, inner front cover, shows that the effect is a packet variant of Charles T. Jordan's “The Alternate Reverse” from his Thirty Card Mysteries, 1919, p. 54, in which cards are alternated face up and face down, and then right themselves magically. The colors of the cards appear to have no bearing on the effect, which is one of cards magically reversing, not of colors separating.

In 1940, Walter B. Gibson published a one-phase color separation effect with a six-card packet: “Like Seeks Like” in The Jinx, No. 91, May 1940, p. 569. Gibson used the Glide to cause the colors to separate. This has been cited as evidence that the effect is older than Marlo's “Oil and Water”, but Marlo argued that important elements differentiate Gibson's effect from his: (1) an emphasized mixing procedure of the colors, (2) repetition of the effect and (3) the metaphor of oil and water to underline the effect.

An unusual red and black separation effect, “Les Seize Rouges et les Seize Noires”, is an early relative of the Oil and Water plot. This appeared in the French conjuring journal Passez Muscade, No. 23, Sep 1924, p. 290, under the byline of “Professeur Magicus”. A deck is examined and shuffled by a spectator and set under the foot of a goblet. The goblet is covered. The spectator next chooses red or black, and all cards of this color are made to travel magically from the deck to the goblet above. The method involves the Mirror Glass principle, a box that switches decks and a prepared set of double-faced cards with red faces on one side and a Deland-like two-card-spread gaff on the other side. When the pack is fanned with the gaffed two-card faces showing, it appears to be a full deck, but when the other side is fanned and displayed, only red cards are seen and they may be counted to prove there is only half a deck left.

It has been pointed out that Samuel Pavloff contributed a Four Ace effect, “What's Up?”, to Hugard's Magic Monthly, Vol. 6 No. 7, Dec. 1948, p. 494, in which, at one point he uses an oil-and-water metaphor, with others, conveying the idea of separation. That is the only similarity to Marlo's “Oil and Water”.

In 1928, Leslie Guest proposed the effect of causing the cards in two banks—one red, one black—to alternate magically. In his column in The Linking Ring, Vol. 8 No. 8, Oct. 1928, p. 639, Guest mentioned that a number of methods could be constructed, and gave as an example preparing a deck by waxing pairs of cards together. This variant effect is now commonly called “Anti Oil and Water”. Marlo reinvented the effect when he published “Oil and Water Climax” in Ibidem, No. 8, Dec. 1956, p. 15.

In Ibidem, No. 15, Dec. 1958, p. 14, Marlo published a variant effect, “The Second B and R Routine”, in which blue- and red-backed cards are mixed and caused to separate, using the card backs rather than the colors of the faces as the focal feature. Marlo also mentions using red-backed Tens cards and blue-backed Jacks. This made the separation seem more difficult since both backs and faces can be clearly differentiated. Some years later, William Miesel published a routine with just a slight change in the effect: Red-backed red cards and blue-backed black cards are used. See “Oil and Water” in M-U-M, Vol. 55 No. 11, Apr. 1966, p. 24. This variation later became known as “Technicolor Oil and Water”.

Shortly thereafter, Karl Fulves published another “Technicolor” routine that exercised greater influence than Miesel's on further variations: “The Cards That Can't be Mixed”, The Pallbearers Review, Winter Folio 1967, p. 140. A few months later (The Pallbearers Review, Vol. 3 No. 5, Mar. 1978, p. 165), Fulves mentioned Ron Edwards's “No Two Ways” in Ibidem, No. 13, Mar. 1958, p. 11, as a related effect. The Edwards trick used a four-card packet, and resembles a puzzle posed to the spectators that has no solution more than it does a magical effect. Therefore, while related, it is not a Technicolor Oil and Water effect.

In the same Marlo article in Ibidem, No. 15 cited above, he also mentions the idea of doing an Oil and Water effect with face-up and face-down cards; see p. 17.

Howard P. Lyons is the first to publish the idea of climaxing an Oil and Water routine by having all the cards change to the same color. Lyons considered the effect an inside joke to show only to other magicians. See “Clown Version” in Ibidem, No. 9, Mar. 1957 p. 180 (of the 1993 book edition of Ibidem, Vol. 1).

Sixteen years later, Karl Fulves reinvented the idea as a serious effect; see “Camouflage” in his Packet Switches (Part Two), 1973, p. 98.

This surprise variant of Oil and Water, in which the four cards of one color change to four Queens as a climax, was originated by Roy Walton and originally published in his Devil's Playthings (1969), p. 15.