CHICAGO (AP) _ Some children fed chloride-deficient formula in the late 1970s later developed some motor, speaking and memory problems, but the study's chief researcher said Tuesday more research was necessary before conclusions can be made about the long-term effects of the formula.

The study, by Dr. Anne Willoughby and her colleagues at the National Institues of Health, examined 20 children who were fed the formula, Neo- Mullsoy, as infants for periods ranging from seven weeks to 35 weeks.

The results appear in the June issue of Pediatrics, the journal of the American Academy of Pediatrics, based in suburban Elk Grove Village.

The formula's manufacturer, Syntex Inc., of Palo Alto, Calif., stopped adding salt, or sodium chloride, to the product from spring 1978 until August 1979, prompted by concerns that the sodium levels were too high.

But in August 1979, Syntex recalled and destroyed 8.5 million cans after reports of illness among infants fed the formula.

Researchers tested the children in 1980 and again in 1982, at ages 2 and 4, for speaking skills, memory and muscle functions.

The tests, formally referred to as Bayley Scales at age 2 and the McCarthy Scales at age 4, each have a mean score of about 100. The total mean of the 20 children participating in the research was also near 100.

Ms. Willoughby said the tests, administered at NIH headquarters in Bethesda, Md., showed that some of the children who were fed the formula longer than others had lower scores on the tests.

But she said none of the children studied ''had dramatic problems. At ages 2 and 4, there's not much they can be expected to do anyway.''

''To link the formula and any long-term problems would be a very big mistake,'' Ms. Willoughby said. ''We need a larger sample. That's what we're working on next.''

Syntex, which stopped making all formula late in 1980, agreed.

In a statement Tuesday the company said, Ms. Willoughby's study ''is not a well-controlled, definitive study. Its findings cannot be extrapolated to any definitive conclusion of potential risk.''

Carol Laskin of Washington, D.C., head of Formula Inc., an organization of 5,000 parents whose children were fed the formula, had mixed reactions to the study.

''In the last two months, we sent out 800 questionaires to parents whose children took the formula,'' Mrs. Laskin said.

''At this point we have gotten a 55 percent reponse rate and 25 percent of those surveyed had their children diagnosed as learning disabled,'' she said. ''The researchers say there is a risk and we agree.''