So says Antonio Neri in his groundbreaking 1612 book of glass recipes, L’Arte Vetraria. Elsewhere he warns in several places not to add “tartar” to any glass destined for yellow pigmentation. Tartar was a common additive to boost the ‘sparkle’ of a glass because it contained a high level of potassium carbonates. These converted to potassium oxide in the melt, which has a higher refractive index than the usual glass flux, sodium oxide.
"Very few people know how to make colors like golden yellow and solid red well. These are difficult and troublesome in the art of glassmaking, since in making them you must stick precisely to the doses, the timing, the details and the materials as prescribed. The smallest error will cause everything to be ruined, and the colors to be irreparably spoiled. Therefore, you must be on guard not to make mistakes. 
Yellow Neon Chandelier, 1995Dale Chihuly.(Columbus, Indiana Visitors Center).
However, the reason for the mismatch is not clear, since Neri also says of his “fern glass,” which is entirely potassium based:
…This frit can be given a wonderful golden yellow color provided there is no tartar salt within, as described in the caution, because then golden yellow will not emerge. This crystal is given to a golden yellow that is far more beautiful and pleasant than can be achieved in cristallo made with Levantine polverino salt and with this crystal unlike the other, every kind of job can be done. “Polverino” was a sodium based plant product used in many of Neri’s glass recipes, which he says was derived from the Kali plant grown in the Levant. The plot thickens when, for yellow, he recommends substituting ‘rocchetta’ another soda based Kali derivative.
His primary recipe for golden yellow is #46, in which he reveals two ingredients responsible for the color: “For every 100 pounds of [glass], add 1 pound of tartar from the dregs of red wine. Use large pieces well vitrified naturally in bottles of wine, because the powder is no good. Crush these raw dregs well, and pass them through a fine sieve. For every 1 pound of these dregs, add 1 pound of prepared Piedmont manganese…”  To this he adds the advice that “the powder is always given in parts and given [to the frit], not to the fused glass, because then it will not tint.”
He also offers advice to add more or less pigment depending on the intended use of the glass: more for thin items, less for heavier ones. “For larger [thick] spit beads, it is said that at Murano they reduce the dose of [wine] dregs and manganese by nearly half.”
For Neri’s lead glass, he uses a different combination, this time pairing copper sulfate with iron oxide: “Take 16 pounds of cristallo frit and 16 pounds of lead calx. Mix them well and pass them through a sieve. To this material, add 6 ounces of thrice cooked copper, made with flakes of the kettle-smiths [chapter 28], and 2 pennyweight of iron crocus made with vinegar [chapter 17].” He goes on to advise, “If it leans toward greenishness, add a little iron crocus, which will remove the greenishness and will bring out a yellow color of the most beautiful gold.
Yellow is one of several colors that iron oxide can form in glass, and is used frequently in low-fire pottery glazes. In that realm, it has a reputation as a difficult, unstable color, as Neri alludes to in his warnings. But in modern, higher temperature borosilicate glass, iron oxide is relied on for a nice yellow. In modern soda-lime glass, cadmium, titanium or the exotic praseodymium are more likely choices. They produced bright reliable color that is stable at the higher temperatures of modern operations. In lead glass, selenium is the modern favorite for yellow.
 Neri 1612, ch. 45.
 ibid, ch 5.
 ibid, ch 46.