Herman H. Goldstine, 90, a mathematician who played a key role in early development of the electronic digital computer during World War II, died June 16 at a retirement community in Bryn Mawr, Pa. He had Parkinson's disease.

Dr. Goldstine, who later worked at IBM, wrote "The Computer From Pascal to von Neumann" (1972), a highly readable account of the history of mathematics and the way it influenced the development of computer science.

During World War II, Dr. Goldstine worked for the Army's Ordnance Department, which had an interest in developing faster and more accurate artillery and bombing tables.

Assigned to the Army's Ballistics Research Laboratory in Aberdeen, Md., he began persuading Army officials to invest money in a computer project underway at the University of Pennsylvania engineering school. Dr. Goldstine became the Army's liaison to the project, which was being led by John W. Mauchly and J. Presper Eckert.

The result, presented Valentine's Day 1946, was ENIAC, short for Electronic Numerical Integrator and Computer.

It was the first electronic digital computer and an unwieldy device -- 18,000 vacuum tubes, filling a room 30 feet by 50 feet and using 150 kilowatts of power. "It was like fighting the Battle of the Bulge to keep it running daily," Dr. Goldstine later said.

The ENIAC could store 20 numbers of 10 digits each in its electronic memory and was a milestone in general-purpose computing. It impressed many at the time by performing rapid digital processing.

Besides his supervisory role, Dr. Goldstine was credited with some of the mathematical underpinnings of the ENIAC. He also said he had a major role in bringing Johnny von Neumann to the ENIAC project after seeing him one day in 1944 at the Aberdeen train station and persuading the math giant to visit Penn.

At the time, von Neumann was attending a scientific advisory committee meeting at the Army's Ballistics Research Laboratory. He was intrigued by high-speed devices that would help with his work on the atom bomb at Los Alamos, N.M. Many of the difficult calculations for the first atom bomb were made with electronic calculators that were essentially office machines.

"Fortunately for me, von Neumann was a warm, friendly person who did his best to make people feel relaxed in his presence," Mr. Goldstine wrote in his 1972 book.

"The conversation soon turned to my work," he wrote. "When it became clear to von Neumann that I was concerned with the development of an electronic computer capable of 333 multiplications per second, the whole atmosphere of our conversation changed from one of relaxed good humor to one more like the oral examination of the doctor's degree in mathematics."

Herman Heine Goldstine was a Chicago native and received bachelor's, master's and doctorate degrees in mathematics from the University of Chicago.

Early in his career, he taught mathematics at the University of Chicago and the University of Michigan.

In 1941, he married Adele Katz, who helped program the ENIAC and wrote an operating manual for it. She died in 1964.

Survivors include his wife, Ellen Watson Goldstine, whom he married in 1966, of Bryn Mawr; two children from his first marriage; and four grandchildren.

After his Army work, Dr. Goldstine worked at the Institute for Advanced Study in Princeton, N.J., helping create a stored-program machine that became the model for the early IBM computers.

He worked at IBM from 1958 to 1984, serving as director of mathematical sciences in research, director of scientific development for the data processing division and consultant to the research director.

In retirement, he spent 13 years as executive officer of the American Philosophical Society, the learned group founded by Benjamin Franklin.

In 1985, President Ronald Reagan presented to Dr. Goldstine a National Medal of Science for "fundamental contributions to development of the digital computer, computer programming and numerical analysis."

He continued writing -- often from his home computer, on which he did his taxes and other tasks.

"In some ways, it's become just another appliance," he told Newsday in 1996. "But it's much better than that because it can interact with you, which is pretty thrilling for children.

"They don't want to hear about old times and vacuum tubes," he said. "They want to push it to the next uses, which is how it should be."

Herman H. Goldstine was awarded a National Medal of Science.