The Perfect Machine (39 page)

From the days of the first failed disk, the ladlers had orders to halt for instructions if a core floated. When the ladling suddenly stopped and McCauley was seen hurrying to the mold and peering through an opened gate, rumors started in the gallery that the igloo over the mold had caved in. As McCauley emerged, his smile of the morning gone, the rumor grew. The word in the gallery was that many of the twenty thousand refractory bricks that had been used to build the “doghouse” were falling into the mold and that the disk was a complete loss.

What McCauley saw through the pour opening in the igloo was a single tapered round core bobbing on the surface of the molten glass. His mind whirled with ifs:
If
only the crowd were miraculously swallowed up.
If
only “we” hadn’t been so cocky with success to think it safe to invite an audience.
If
only one core had floated up on the 120-inch disk to demonstrate the weakness of the anchors.

As he stood back from the heat of the open gate, the ifs gave way
to whats:
What
will the reporters say on Monday morning?
What
damage will result to Corning’s prestige?
What
actions should be taken to save face?
What
will be the effect on the decision of the Observatory Council to continue the program?

The decisions were all his. McCauley ordered the ladlers to resume ladling. He would fill the mold and salvage the disk by whatever they had to do, even if it meant laboriously grinding pockets in the back face. He quickly decided on a triage for his own time. He would need all his ingenuity to salvage the disk. The press release, and the challenge of keeping reporters and photographers away, was a problem left to the ingenuity of Quigley and Amory Houghton.

As the ladling went on, twenty-two cores broke loose, slowly floating to the surface of the semimolten glass. As each core popped loose, the vacated area would fill with molten glass. Wilson and Ruocco, the ladlers, filled and poured more and more ladles. The day stretched on, longer than McCauley anticipated. The crews were tired. Wearing street shoes like the rest of the crew, Wilson slipped and banged his head against a ladle. He was knocked unconscious, though he recovered in time to finish the ladling.

It was six o’clock in the evening before the last ladle of glass was poured into the mold. Most of the invited guests had already left. The line of more than six thousand spectators who trooped through the gallery had dribbled down to a trickle. Only those who were nearest the mold had any inkling what had actually happened.

Quigley and Houghton had already prepared press releases and interview comments, which the newspapers and broadcast reporters dutifully reported. “A trifling accident which temporarily suspended the work,” the release read, “had occurred during the pouring.” Walter Adams, the ranking representative from Pasadena, was quoted to the effect that the cores that had broken loose would not affect the disk. The release was issued quickly enough to be picked up for the early-morning newspapers.

McCauley knew the truth. That evening, when the workers and the crowd went home, he had to deal with twenty tons of molten borosilicate glass that had twenty-two cores of refractory brick floating on its surface like toppings on a sundae.

When the factory finally emptied, the melting-room crew was standing by, ready to move the filled mold from under the heated igloo to the waiting annealing oven. With the crowd and reporters gone, McCauley had experienced tank men don protective asbestos aprons and face masks, and try to remove the floating cores from the surface of the disk with long-handled grappling tongs. The glass had cooled enough from the original 1500°C melting temperature to become a viscous, heavy syrup. After a few minutes in the extreme heat of the open gates to the igloo, the men with the tongs would stagger back for breath. It was impossible to remove the cores.

The next attempt was to use long iron rods to break off the tops of the floating cores. Perhaps, McCauley hoped, the remainder of the cores would be embedded in the surface no deeper than the layer of glass that would ultimately be ground off the disk in making the concave surface. The effort to break up the cores was more successful than the effort with tongs, but the surface was far from a pretty sight. This wasn’t the perfect disk the telescope needed.

By ten that night the crews had broken up as much of the cores as it seemed possible to reach. The glass and core pieces stood proud of the original mold height, so work crews had to build up the perimeter of the mold with insulating brick, lest the heated cover of the annealing oven press down on the cores and further embed them in the glass. With the new row of bricks in place, McCauley allowed himself a break while the mass of glass slowly cooled down to the temperature at which the long controlled cooling of the annealing would begin. It was his first break since the operation had begun at eight in the morning.

He hadn’t eaten since before the pouring began. Hostetter, who had spent the afternoon and evening with visiting VIPs, joined him at the Athens lunch counter, a popular Corning spot. The room was empty except for the waiter. McCauley, famished, ordered a roast beef sandwich.

“Sorry gentlemen, no beef.”

“Ham and eggs?” McCauley asked.

“No ham.”

“Bacon and eggs?”

“No bacon.”

The crowds, more people than Corning, New York, had ever seen, had eaten everything. The two men settled for eggs and coffee.

When McCauley got back to the blowing room, the disk was cool enough for its trip to the annealing oven. He pushed a button and the screw hoist slowly lowered the mold and glass below the floor. Crewmen on the hand-cranked windlass moved the heavy load along the tracks until it was centered beneath the annealing kiln. McCauley thought to himself that in an hour the disk would be inside the oven, hidden from the world. The long day would be over.

He pressed the button on the hoist, the screw hoist slowly turned, and the disk started upwards. It was halfway on its journey when the overload switch opened and the hoist stopped. With the switch reset, McCauley tried lowering it a short distance, then raising it again. The jammed lift wouldn’t budge. Sunday night was no time to call the Whiting Corporation in Chicago Heights, Illinois, for service.

To maintain the temperature of the disk, McCauley and the crews filled in the open space between the top of the mold and the bottom of the annealing oven with Sil-o-Cel. The temperature controllers for the annealing oven were set to hold the interior at 500°C to postpone the
cooling as long as possible. It was all they could do. McCauley and the exhausted crews went home.

The early editions of the morning papers were already out. The big headlines and photographs of the pouring had crowded the usual stories about Mussolini’s alliance with Austria and Hungary, and Japanese troops in Manchuria, off the front page.

He had told them not to wait up, but McCauley’s whole family was there when he got home Sunday night. They had read the late edition of the
Evening Leader
and heard a stream of rumors about the disk. They remembered how ebullient he had been in the morning, and through most of the day. A quiet man, he had been thrust into the limelight and had enjoyed it. Now his face was gray with exhaustion and frustration, as he sat down at the dining room table where he had worked so many evenings while his children did homework.

“I know what happened,” he said. “We just have to see that the bolts won’t do that.”

18
Salvaging Hopes

Even before Walter Adams and Francis Pease brought home the news that the two-hundred-inch mirror had been successfully cast, the barrage of publicity from Corning galvanized Pasadena. No one had been willing to admit it openly, but ever since the dark days of the GE experiments, much of the project had been held back—partly out of fear that the telescope
couldn’t
be built and partly to avoid the inevitable publicity and questions that would follow each major decision for the project. After the hoopla in Corning, it was inevitable that questions would be directed to Pasadena: How long will it take to finish the telescope? How far will the telescope see? Where will the telescope be sited? What will it look like? Who will get to use it?

Hale and Anderson had half answers for the reporters’ questions. When Hale had said in 1928 that the telescope might take five or six years to finish, the astronomers and engineers close to the project thought him excessively pessimistic. Six years later, if pressed, Anderson and Hale would suggest 1939 as the date when the telescope would be finished, but the date was made of whole cloth. The real answer, which reporters wouldn’t welcome, was that they didn’t know how long it would take to grind a mirror, to build a telescope, or to put together an observatory. They hadn’t announced where it would be sited because the Observatory Council was reluctant to sign leases or even announce a site decision until they knew they could build the telescope. Fundamental questions about the design of the telescope hadn’t been agreed on, and no plans had been made for the mount and observatory, because the Observatory Council and the various design committees hadn’t been convinced that they would be able to build the telescope. And while the original proposal to the IEB had specified that the Carnegie Institution would work closely with Caltech in the design and ultimately in the operation of the facility, until they were reasonably sure that the telescope would be built, no one from Caltech or the Observatory Council wanted another round of negotiations with John Merriam.

It would be months before anyone knew how the disk would emerge from the annealing process, and whether it could be salvaged, but the Corning officials and Walter Adams were convinced that the success of the ladling process meant that a mass of glass as large and as complex as the two-hundred-inch disk
could
be cast. Hale wrote to Arthur Day expressing his gratitude and admiration for Corning’s achievement: “If the mirror proves a success, it is not likely to be repeated very soon.” Privately McCauley, Houghton, and others at Corning assured Hale that if the disk couldn’t be salvaged, Corning would pour another.

With the disk seemingly assured, it was time for decisions.

The committee on site selection had been accumulating data for years. John Anderson, the de facto chairman of the committee, used his own old-boy network to call on the geology department at Caltech for help.

Geology at Caltech had begun with a Carnegie Institution program, headed by Harry O. Wood, who had worked for Hale’s National Research Council during the war. Anderson collaborated with Wood to develop a seismometer sensitive enough to detect small local shocks. Charles Richter (of Richter Scale fame) and Gottingen-trained seismologist Beno Guttenberg later came to Caltech as the core of a geology faculty, and the new Department of Geology absorbed the Carnegie Institution Seismological Laboratory. When it was time to assess potential sites, Anderson turned to his friends at the Geology Department, who were using the successor instruments to the seismograph he had helped develop.

A team of geologists, headed by J. P. Buwalda, assessed the sites on the shortlist and reported that Horse Flats, north of Mount Wilson, was too close to the San Andreas Fault. They found no problems with Table Mountain, east of San Bernardino, or with Palomar. The long-term seeing-test data from Table Mountain and Palomar were roughly equal, but Hale had favored Palomar from the beginning, even when there were some indications that under the best of conditions, the seeing at Palomar did not equal the remarkable seeing at Mount Wilson.

In March 1934, Hale, Adams, and Anderson traveled up to Palomar with Dan Tracey, the forest ranger at Mount Wilson, in the Pierce-Arrow touring car the Mount Wilson Observatory had bought before Einstein’s first visit. Hale had been up the mountain before, and Anderson had been several times to check the monitoring equipment on the mountaintop. This time the three of them were looking for a location to site the biggest telescope in the world.

The mountain had changed since 1903, when Hussey wrote the report that captured George Hale’s imagination. Once a source for timber to build missions, by the turn of the century Palomar, within a few days’ stage of Los Angeles or San Diego, had become a popular summer
resort, with seasonal hotels and a tent city that blossomed each summer in Doane Valley. The trail that ascended the west shoulder of the mountain, called the “Nigger Grade”
^*
by the locals, and only later renamed Nate Harrison Grade, took a full day for a team, so there were halfway camps where travelers could rest before the ascent. Those who survived the deerflies in June, mosquitoes in July, and no-see-ums in August could enjoy the unspoiled splendor of acres of summer ferns and azaleas; meadows of dense grass, wildflowers, and butterflies; and the forests of big-cone spruce, white fir, and California black oak.

The automobile brought distant resorts in range of a one-day drive from Los Angeles and San Diego, putting an end to the Palomar resort era, although Bailey’s Palomar Lodge held on. A few intrepid souls liked to challenge the switchbacks and hairpin turns of the Nate Harrison Grade in automobiles. The grade took its toll in gears and overheated engines. The real challenge was the downhill run. The favorite technique was to tie a large tree to the rear bumper and drag it downhill; the right-size tree would provide the perfect brake. At the bottom the local Indians took the trees for firewood. It wasn’t a road built to haul telescopes up a mountain.

By the 1930s cattle ranching had taken over much of the mountain. The Mendenhall Ranch, on the southern side of the mountain, was the largest. Mendenhall had been around long enough to work through breeds, from mixed-breed, white-faced “California cattle” to Herefords and finally to polled Aberdeen Angus steers that thrived on the high-meadow grass. To the northwest, a sheep-farming effort by a Frenchman named Foussat had failed, leaving behind the name French Valley. Locals attributed the failure to locoweed. There were smaller ranches, including some failing efforts at apple orchards. William Beech and his wife kept a cabin high up the slope in the French Valley. Beech had run a weather station and monitored some instruments for Caltech for five years.