When Mrs. Thatcher talks, interviewers interrupt -- why?

According to three research psychologists, British Prime Minister Margaret Thatcher has an unusual distinction. When interviewed, she is typically interrupted more frequently than other senior politicians.

The reason, they say, is not that interviewers are rude. Mrs. Thatcher appears unconsciously to signal a readiness to yield the conversation when, in fact, she means to go on talking.

This emerged from a detailed analysis of a Thatcher TV interview, an analysis made by Geoffrey W. Beattie of the University of Sheffield working with Anne Cutler and Mark Pearson of the University of Sussex. Their study is part of the fairly new research field dealing with conversational cues and signals.

These include cues as to when a speaker is or is not ready to let the other person have a turn. Cues are given by the pitch, timing, and intensity pattern of speech. Gestures, gaze, and other body language also are important.

Starkey Duncan Jr. and Donald W. Fiske of the University of Chicago, who have pioneered in this research, have noted that no one cue, particular type of cue, or grouping of cues seems basically more important than any other. Instead, they have found that the turn-yielding signals have a breadth and versatility that make this catalyst of conversation quite flexible.

This flexibility also opens the possibility of ambiguity when the cues are given in what might be considered an unconventional way in a specific situation. This happens in the Thatcher interview.

It was conducted by ITV (Independent Television) interviewer Denis Tuohy on a program called TV Eye in April 1979. Describing their study in Nature, the three psychologists explain that they used 40 extracts from the interview. These included instances where Thatcher meant to yield the turn as well as where the turn was disputed.

Students, acting individually as judges, then assessed these extracts in video, audio, or printed form. They agreed that many of the interruptions occur at points where Thatcher appears to have finished her turn when, in fact, she had meant to continue.

The cues given include such things as dropping the pitch of her voice. Both the level to which it dropped and the speed with which it dropped seem to be important. The researchers suggest that she may unconsciously use such cues in ways the interviewer doesn't expect.

For example, they note, ''If she considered that her most decisive cue . . . was letting her voice drop to around 140 Hz (Hertz) instead of keeping it no lower than 160 Hz whereas her interlocutor considered that her most decisive cue was a rapidly executed final fall (in pitch) rather than a slow fall . . . (confusion over whose turn it was to speak would occur in) precisely those cases which were disputed. . . .''

Conversation is a skilled performance. As the researchers observe, ''Successful manipulation of these very subtle turn-taking signals must certainly rank as one important aspect of this . . . skill.'' New hope for jojoba?

Jojoba - the much touted ''wonder'' plant whose nuts yield high-quality oil - has been a commercial dud. But now the possibility of raising fields of genetically identical plants opens a new prospect for eventual success.

The problem is that only some female plants are good oil producers. It takes about three years of growth for jojoba (pronounced ho-ho-ba) farmers to tell male and female plants apart. Further time is needed to sort out the high-yielding plants. Given this long delay and the fact that there is no ready market for the oil, commercial growers have not taken to the plant.

Now at Twyford Plant Laboratories in England, a branch of International Plant Laboratories, jojoba is being propagated vegetatively. The tips of shoots or roots are cultivated, The ensuing plants can then be slipped to generate still more plants. In this way, whole fields of genetically identical plants, selected for high yields of oil, can be grown.

The new technique is to be field tested in Arizona. If successful, it could make jojoba oil easy to obtain. It would still remain to be seen how large a market would develop for the oil. But a major barrier to its commercialization would have been breached. Music of the spheres

Scientists revere 16th-century astronomer Johannes Kepler for discovering the laws of planetary motion. But his mystical notion that God created the universe to be beautiful and simple, which he used to account for his laws, inspires only amusement.

Judith Field of Britain's Science Museum suggests that Kepler's notion deserves more respect.

Kepler explained the position of planetary orbits in the solar system as representing a set of nested polyhedrons (a cube, a tetrahedron, and so forth). According to New Scientist, Field has shown that this theory gives the orbits of the six planets Kepler knew (out as far as Saturn) with an error of only a few percent.

Kepler was also fascinated by what he considered the harmony, or music, of the planetary spheres. In this theory, Saturn's motion defines a pitch with which the motions of the other five planets are harmonized as corresponding to musical notes within a single octave.

Field shows that, when Kepler's geometrical theory is refined to conform to this musical requirement, the error in its planetary predictions is less than 1 percent. That's within the limits of the observations Kepler had available.

Prediction within observational accuracy qualifies a theory for serious consideration. Moreover, Kepler's mysticism is no more weird than modern quantum theory, which considers a material particle to exist only when it is observed.

While noting that Kepler's theory won't hold water today, Field urges modern scientists not to smirk at its weirdness. For its time, it was a bold and apparently accurate theory. Volcanic dust and climate

Dust from El Chichon's eruption in Mexico last March 28 will become a global veil this year and next. It should then have its strongest climatic effect, if any. What might this be?

Climatologist J. Murray Mitchell of the US National Oceanic and Atmospheric Administration says it may tilt the odds slightly in favor of cooler weather. However, he explains in Weatherwise that meteorologists can't predict exactly what will happen.

He notes that ''. . .the weather of a season is shaped by many factors having little to do with volcanoes. . . .'' It results, he says, from ''a kind of endless shoving match that goes on between the atmosphere, the oceans, and the polar ice and snow.''