Most types of cancer are still on the increase, some drastically, a National Cancer Institute official told a Senate health subcommittee yesterday.

Among men, eight of 10 major types -- including bladder, prostate, lung and intestinal cancers -- are increasing in number. Among women, eight of 13 types -- including lung, uterine, breast, bladder and kidney cancers -- are rising.

And among both white men and white women, one of the most rapid increases is in melanomas -- frequently fatal skin cancers -- mainly because of "our interest in sunshine and a good tan, in short, the fact that we're sunworshipers," testified Dr. Marving Schneiderman, a cancer institute statistician and epidemiologist as well as associate director for scientific policy.

But he also said melanomas may be on the increase because more of the sun's powerful ultraviolet rays are reaching earth with a deterioration of the ozone layer. Some scientists think this protective layer of molecules far above the earth is being degraded by man's chemical activities.

Stomach cancer is down sharply in both men and women. Leukemia incidence is also down. So is the incidence of cancer of the female cervix, the neck or entryway to the womb.

Here, Schneiderman said, wide use of the Pap smear has enabled doctors to catch changes before they become cancerous.

He reported other optimistic facts, too. Though the incidence of many cancers -- and cancer in general -- is on the increase, the rise has begun to level off in the past five years or so.

There have been large gains in survival for victims of a few cancers, especially the leukemias and related lymphomas. In recent years there also have been significant gains in survival in some of the more difficult-to-treat solie tumors, including breast cancers, he said.

There has been a "substantial" reduced incidence recently of breast cancer in women under 45, he said, possibly because more women have been bearing babies early. Statistics show that women who bear their first child early develop breast cancer less often than women who bear late.

Here -- as in many other cases -- he had to cite the difference in child-bearing as a "possible" cause.

"We don't have very much exact information about causes," he told subcommittee Chairman Edward M. Kennedy (D-Mass.).

But "of course" cigarette smoking is a major cause, Schneiderman said. He spoke of many other causes related to "the way we live" -- for example, that young women who have many sexual exposures to different men develop cancer of the cervix more often.

He spoke of the ubiquitous chemicals in the environment and work place, and the "alarming" fact that "we are adding to the number very fast." But also, he said the fact that the annual increases in cancer incidence and mortality "have been starting to level off" indicates "progress" against some chemicals and that "American industry has clearned up" in part.

Also, he said, the lung cancer incidence rate has started dropping among younger and middle-aged men, as relatively fewer are smoking after age 34. At the same time lung cancer is climbing sharply in women, as more girls and young women smoke.

Schneiderman cited another striking case of almost certain cause and effect. Cancer of the endometrium, the lining of the womb, saw a sharp rise after women started wide use of estrogens after menopause. After wide publicity about these cancers, both use of estrogens and incidence of endometrial cancer dropped.

"This has important implications for both direct prevention and basic research," he said. "It tells us that estrogens very likely are cancer 'promoters.' and that the result of interfering with a cancer promoter can be seen in a very short time. We have a lot of looking to do, but this is encouraging."