688 In these sentences the interrogatives with -ti.xa are not followed by uka resumator. In non-Missionary Aymara -ti.xa occurs infrequently, as a subordinator of verbs and demons- stratives. It may also occur on interrogatives, but when it does it is usually ordered first in a sentence and followed later by uka resumator, as in the examples given in 7-4.24. An example of code-switching in which Spanish all but submerges the Aymara is the following sentence, again from the Baptist sermon referred to earlier: Yo creo que jani.w hermana.naka.s. I think that no ~ sister ~~ 'T think not, sisters.' The only Aymara in the above sentence is the negative particle jani with the absolute final suffix -wa (reduced to w) and the Aymara plural suffix -naka, which is followed by the Spanish plural -s. The meaning conveyed is some- what stronger than the above translation would imply, due to the presence of -wa; a closer approximation would be "No, sisters, I know it's not so.’ Although successive embedding in noun phrases is possible in Aymara (see 7-4.2), it is not very common. It occurs with more than usual frequency in a story told by a Protestant minister from La Paz (Wexler 1967:454-456).