I will not look down. I’m sitting on a metal-pipe scaffold perched on a lip of roof. Behind and below me are 128 feet of perfectly clear autumn air. In front of me is some of the most intricately decorative concrete ever cast. The Sears Tower, 14 miles south, lies blue and vague on the horizon. I will not look down.

I am up here, just below the dome of the Baha’i House of Worship in Wilmette, to see the concrete restoration project being undertaken there. Next to me, an unfazed Bob Joyce is holding two six-inch squares of concrete up against the wall. He’s general superintendent of William Hach and Associates, the concrete restoration contractor that erected the scaffolding. It seems that one square matches the temple wall perfectly in color, the other is a perfect match in texture, so if you can imagine them together. . . .

On this part of the temple’s south-facing wall, the 52-year-old decorative concrete is cracking up. Several of the dentils (small rectangular projections that in fact resemble teeth) below the cornice have eroded back to ragged diagonals, exposing the rusty ends of reinforcing steel bars. Deep cracks run diagonally down into the once-sharp upper corner of the horizontal molding above the arched windows. Here and there splotches of black and white mar the surface. The white “efflorescence” shows that water has passed through the concrete shell, dissolving minerals on the way and then depositing them on the surface. The black deposits are just dirt.

If you take the el all the way north to Linden Avenue in Wilmette, then walk a few blocks east to the lake and gaze from a sidewalk at the architectural tour de force that is the House of Worship, you won’t see any of this deterioration. The decaying parts are too small and too high up to detract from this massive yet ethereal building that looks as if it had been cast from fine lace. No concrete is about to come tumbling down, nor has the temple’s structure been affected. But NSA Properties, Inc.–which is the property-management arm of the National Spiritual Assembly of the Baha’is of the United States–is determined to put the structure to rights. “Most buildings,” says Ronald Precht, spokesman for the Baha’i National Center in Wilmette, “people look for them to last a while and then they’ll move. But this building is so important to Baha’is–both historically [it was placed on the National Register in 1978] and because the son of the founder of our faith laid the cornerstone [in 1912]–it’s more like, say, the Washington Monument. It is the physical manifestation of the spiritual teachings of the Baha’i faith.”

So the temple, which took more than 40 years to erect and was not finally dedicated until 1953, is already being restored. Bob Joyce does not find this surprising. “The average building we work on is 15 to 20 years old,” he says. “I’ve seen 10-year-old buildings in worse shape, . . . due not to bad design but to poor workmanship. I’ve been on buildings downtown where the original forms won’t fit because the building is so crooked.” Crookedness is certainly not a problem with the temple, but diagnosing its current ills has so far involved clambering on scaffolds, combing through old photographs and drawings, drilling and removing 14-inch-deep cores from the concrete, ransacking the memories of a man who worked for temple contractor John Earley (“the man who made concrete beautiful”), picking up and sieving a bucketful of pebbles eroded from the wall to check their size, and planning a miniature mountain climb up the temple’s dome.

Before the temple, of course, there was the Baha’i faith–a set of beliefs in the unity of God, mankind, and revealed religions–founded in 1844 in Persia. The new religion made its first U.S. appearance in the garbled report of a Presbyterian missionary to the World’s Parliament of Religions held in Chicago in conjunction with the 1893 World’s Fair. It found its first U.S. converts here a year later.

Not surprisingly, it was the Chicago Baha’i community, the oldest in the United States, that took up the suggestion of a temple, petitioning the leader of the faith “that in these parts and regions there may arise a Mashrak-el-Azkar, built in the Name of the Glorious God, and that there may go forth from the shelter of its beauty, rays of brilliant light.” The Chicagoans likewise spearheaded the selection and purchase of the Wilmette site, seven acres of high ground between Lake Michigan and the Metropolitan Sanitary District’s North Shore Channel. In 1912, the white-bearded son of the founder of the Baha’i faith came to Wilmette. After laying the cornerstone for the House of Worship, ‘Abdu’l-Baha straightened up and declared: “The Temple is already built.”

Despite this assurance, the fact that Baha’is were neither very numerous (fewer than 2,000 in the U.S. in 1920) nor very wealthy meant that construction proceeded in slow stages: the foundation (including meeting rooms) by 1922, the plain above-ground structure and dome in 1931, the exterior ornamentation by 1943, and the interior ornamentation and landscaping by 1953.

In 1912, not only was the temple not built, it was not even designed. The Baha’i faith dictated only that it have a dome, that it admit natural light, and that it have nine sides. These features are meant to reinforce the temple’s symbolism of God’s and mankind’s unity. Fifteen architects submitted designs to the 1920 Baha’i Temple Unity convention, which chose the “Temple of Light” of Louis Bourgeois, a French-Canadian architect of mystical bent and a Baha’i convert. In Bourgeois’ design, the Italian Renaissance dome covered with quasi-Islamic arabesque tracery served also as a skylight, admitting natural light by day and allowing the light of the Baha’is to shine forth by night. Its decorative interwoven detailing continued to ground level in curves and scallops that Bourgeois saw as echoing the “beautiful spiral curves which the heavenly bodies trace in the sky.”

Just how to capture those curves on an actual building was to become quite a problem, but one that did not have to be solved right away: the Baha’is’ financial constraints dictated that the plain, concrete-and-steel structure be built first and the ornamentation added much later. Bourgeois had wanted to blend many architectural styles into one “unity” in the temple, but clearly the unadorned Bauhaus style did not interest him. As the structure rose–so far resembling nothing more than a huge concrete-and-steel mushroom–on the Wilmette lakeshore, Bourgeois slaved away in his Chicago loft on the exterior ornamentation, rendering some of the largest architectural drawings ever made. “Occasionally as the [drawings] progressed,” writes Frederick W. Cron, “he would climb up on a tall stepladder and view [them] through binoculars to see [the design] as a whole.”

A Baha’i Materials Committee spent the 1920s considering candidates for the decorative superstructure–limestone, granite, marble, terra-cotta, sandstone, tile, brick, cast iron, aluminum, bronze, and concrete. Narrowing the list, they commissioned small test sections of stone, terra-cotta, concrete, and cast aluminum, and left them on the grounds for years to check for weathering. But when in 1932 the committee was ready to proceed, only one producer had the nerve to bid on such an intricate job–concrete craftsman John J. Earley of the Earley Studio outside Washington, D.C.

But isn’t a “concrete craftsman” in the same fix as a classical musician with a kazoo? The medium may allow for proficiency–but art? Concrete itself is a carefully proportioned mixture of three sizes of particles. Cement, the finest, is a manufactured powder that forms a paste when wet, gluing itself and adjacent materials firmly together as it dries. Sand (“small aggregate”) and various kinds of rocks and pebbles (“large aggregate”) add bulk and strength, and at a lower cost than cement. Concrete’s great advantage is of course its flexibility: unlike stone, which must be found or carved into shape, concrete can be mixed and poured into molds (“forms”) of any design; this can be done either at a central factory (“precast”) or at the spot where it will be used (“on-site”).

In short, concrete is strong, gray, and cheap, but it might seem an odd choice to embody Bourgeois’ delicate, fanciful designs. H. Van Buren Magonigle, an architect with a mind as stuffy as his name, surely thought so. He told the 1920 Baha’i convention that concrete was “the most repellant object imaginable. . . . Re-enforced concrete does not weather at all, it merely gets dirty, and it has no beauty of surface, it has no translucence of surface, and it is an exceptionally ugly color. It is almost impossible to get anything but an ugly color, and if it is painted it looks worse than it did before.”

Before the middle of the 19th century, concrete had been used only to hold other materials together; but by about the time the Baha’i faith was being born, it began to be used as a building material in its own right. Improvements in the late 19th and early 20th centuries made it stronger, cheaper, and more reliable: steel reinforcing bars complemented concrete’s great strength, which is to resist compression; and manufactured portland cement gave concrete a consistent strength–and the consistently gray color of limestone and clay. But despite an effortful public-relations campaign, concrete still was considered an eyesore that people of taste would prefer to disguise as something else. Hence, when Chicagoan George Frear patented a concrete block in 1868, he called it “Frear stone.” According to William Coney and Barbara Posadas, authors of an essay called “Concrete in Illinois,” the manufacturers of precast concrete block “advertised it as elegant, inexpensive, durable, convenient, fireproof, and resistant to wind and weather.” So far, so good–“but their reluctance to acknowledge the material as concrete combined unfortunately with the obviously artificial look created when all ‘stones’ in a building were cast in the same pattern.” In a more radical and perhaps more successful attempt at disguise, the architects of the Edgewater Beach Hotel had concrete ceilings poured into forms made of undressed lumber wetted so that the concrete would show the grain; the concrete “beams” were then stained the color of wood.

John Earley, a Washington, D.C., plaster and stucco contractor, took a different approach–looking deeper into the material itself rather than into what it could be camouflaged as. If the cement’s gray color is unappealing, why not let the large aggregate show through? Earley hit on this notion about 1916. In 1934, by which time he was identified as an “architectural sculptor,” he was explaining it to the American Concrete Institute this way:

“We reasoned that if every particle of stone exposed upon the surface of the concrete might be considered as a spot of color in juxtaposition to other spots of color, all the knowledge of color and texture of the mosaicist and of the pointilist painter could be immediately applied to concrete. . . . It would, if a technique could be devised, permit concrete to participate in the traditions of these older arts and transform it almost immediately into an acceptable architectural medium.”

Technique was essential, but in those early days it was also labor-intensive. Before the invention of chemical retarders to delay the setting of cement, Earley’s workmen had to remove the forms before the concrete had fully set (while it was “green”) and remove by hand the surface sand and cement with wire brushes and muriatic acid. Earley first tried this, with dramatic results, on a gray park wall being planned for a Washington, D.C., project: when the aggregate–Potomac River gravel–was exposed, the wall took on a glowing, creamy tan color. Earley was thus spared the trouble and expense of applying a cosmetic layer of plaster or stucco over the concrete, as was usually done.

But the newly visible pebbles did sometimes bunch up, giving the wall a slightly blotchy appearance. The problem turned out to be the common practice of using “uniformly graded” aggregate, with roughly equal portions of several sizes of particles from sand-size on up. (This is still common practice, as in the exposed-aggregate sidewalk on North Michigan south of the Chicago River.) Through patient experiment, Earley found that he could get a consistently uniform appearance if he used a step-graded aggregate, in which the pebbles (large aggregate) were all about the same size, and the sand particles (small aggregate) were all about one-tenth that diameter.

The Earley Studio began doing more work in concrete, and Earley himself became a passionate partisan of his chosen building material: “Concrete is beautiful,” he wrote again and again, arguing against the H. Van Buren Magonigles of the world, and “beautiful concrete is economical. . . . [It] should be in the hands of all architects because with concrete they may bring into reality ideas and dreams that have wanted only the proper medium in which to execute them.”

So it happened that Louis Bourgeois had come to the right place when he first stopped by the Earley Studio in 1920 with a photograph of his nine-foot-tall plaster model of the Baha’i House of Worship. “It soon became clear,” Earley wrote later, “that this Temple was the dream of Mr. Bourgeois’ life . . . the symbol of a new religion in a new age.” The two discussed the project, then and later, but Earley didn’t actually get the ornamentation job–his studio’s largest ever–until two years after Bourgeois’ death in 1930. By then the gray structural concrete of the temple, and the steel and glass of the dome, were in place. It was left to Earley, his partner Basil Taylor, and their crew to give Bourgeois’ dreams and sketches of the ornamentation material form.

They began by building a full-size wooden replica of one section (one-ninth) of the temple dome at Earley’s new plant in Rosslyn, Virginia. Earley soon realized that the dome ornamentation could not possibly be cast in one piece. Such a creation “would tear itself to pieces in its first drying.” He divided the dome into 388 sections, each section to be created through a laborious process: a part of Bourgeois’ drawings was to be redrawn onto modeling clay, the clay model used to make a plaster mold, and the plaster mold used to cast a plaster model. After the plaster model was checked for accuracy on the one-ninth replica, it was used to make another plaster mold into which the final concrete mix was poured. (Some sections could be duplicated, so that in the end only 24 molds were required.)

The aggregate for most concrete is usually whatever gravelly stuff is most common in the immediate area. It doesn’t pay to ship large amounts of bulky generic rocks very far; most of them will just support the weight of and take up space inside the walls of, say, a parking deck. But in the case of the temple, Earley had two unusual requirements. First, it was supposed to last a long time. Second, it was–again according to Bourgeois’ dream–to be as white as possible without being a dead, chalky, plaster white.

“To get this effect,” writes Cron in his biography of the contractor, “Earley chose a white opaque quartz . . . that would reflect light from its broken faces, and mixed with it a small amount of clear translucent quartz [about 30 percent] to provide brilliance and life. From certain angles the particles of clear quartz looked like dark specks on a white background, but from slightly different angles they reflected light like tiny mirrors. . . .” They still do, and the resulting quartz-aggregate concrete was tougher than most other kinds of concrete and many kinds of stone.

But could it hold up its own weight? When Earley’s workmen took the first concrete casting–60 square feet, no more than six inches thick, full of holes, and weighing close to three tons–out of its forms and began wire-brushing it to expose the quartz, it cracked ominously. It might have been strong enough if left to cure fully in the form, but then the twinkling quartz would have been concealed in a featureless gray-white wall.

The studio had solved a similar problem before, when casting Lorado Taft’s “Fountain of Time” statue on the Midway near the University of Chicago. In that case, the concrete had to be mixed very soft (with lots of water) in order to flow smoothly into all the nooks and crannies of a complicated form; but it also had to be very stiff (retaining little water) so that it would be strong and not shrink away from the walls of the form when drying. Earley had solved this dilemma by pouring the concrete wet and then, once it had filled the form, placing rags and burlap on the exposed surface to draw out the excess water by capillary action–thus stiffening the concrete in place. But now, working on the Baha’i dome, they had already done this, had already extracted all the water they could–yet they needed to get still more out so that the green concrete wouldn’t crack when it was taken out of the form.

Earley’s solution was an elegant bit of applied physics. Larger particles, with less surface area per unit of volume, exert less of a pull on the water around them–so that capillary action can pull more water out. So Earley increased the average diameter of the sand by 15 ten-thousandths of an inch, replacing a .0125-inch sieve with a .0140-inch. It was enough. “Considered casually,” Earley wrote with pardonable pride, “it seemed ridiculous that so small a change to but one of the ingredients should make so great a difference in the character of the concrete.”

With the same attention to detail, the Earley Studio turned out all the panels for the dome, then the decorative concrete for the clerestory below the dome and the two lower levels during the next ten years. World War II delayed the interior decoration work; Taylor finished this after Earley’s death in 1947. “Every precaution has been taken to make the concrete as well as it can be made in the present state of the art,” wrote Earley in 1934. “We believe that the concrete in the Baha’i Temple will endure better than terra-cotta, freestone, marble or any other building stone excepting granite.”

Earley knew that even his quartz-aggregate concrete would need maintenance–he designed the dome, for instance, so that any one panel could be removed for repair or replacement without disturbing the rest–but he may not have expected it would need maintenance so soon. In 1934, the very summer that the dome was finished, dust storms from the drought-ridden plains imparted a brownish tinge to the new concrete. (Early photographs show that the newer, lower layers of decorative concrete were noticeably whiter than the dome.) A 1973 cleaning helped restore the color. But some “maintenance” is actually destructive: the outside temple stairs have suffered severely from the use of salt to remove ice and snow.

In the fall of 1983, the Baha’is got serious about maintaining the temple. A weekend seminar of 25 Baha’i volunteers–professional engineers, architects, and contractors–gave the House of Worship a thorough going-over, and they minced no words in their report to the National Spiritual Assembly: “The present condition of the building, though serious and critical, is not actually disastrous at this time”–but without immediate action it soon would be. The biggest culprit: leaks. According to Robert Armbruster, an affable civil engineer who was at the weekend seminar and is now general manager of NSA Properties, it then took 18 months to do only the most urgent repair work: stop the water from coming through the glass dome, the flat portions of the roof, and the gutter that runs around the cornice at the base of the dome. By 1985 it was time for Armbruster and various consultants to assess the damage to the ornamental concrete–and to begin a trip into the institutional memory of the temple construction itself.

“We were trying to figure out why we couldn’t find any rust stains in the joints of the dome,” since the concrete sections are held in place by metal bolts, says Armbruster. “Bob Shaw’s mother-in-law [Shaw, a builder, was another attendee at the seminar] had a booklet dating from the 1930s about the temple, and it referred to “copper-bearing steel.’ It turns out that that was an early form of stainless steel,” specified by Earley precisely to avoid the rust that the engineers expected. In another discovery, the consulting engineer who ascertained that there was no structural damage to the building also found that, if the “copper-bearing steel” bolts holding a dome panel had failed, Earley had provided a second support mechanism–a panel “could fall only about one-eighth of an inch” before the backup supports caught it.

The fact of such rediscoveries is not surprising; what is unusual is that there haven’t been more. “Most typically on a 50-year-old building, we have no information, or very little,” says Jack Stecich, an IIT-trained engineer and project manager for Wiss, Janney, Elstner Associates. Earlier this year, NSA Properties hired Wiss, Janney–a nationally known, Northbrook-based firm that specializes in evaluating buildings (and building failures)–to look into both the building and its archives, and to devise a diagnosis and treatment plan for the House of Worship.

Stecich found the Baha’i institutional memory remarkably complete. (Religions, seeking to build traditions for eternity, tend to save lots of documents; this tendency is compounded in those, like Bahaism, that have originated in our modern Age of Documentation.) The layman might be impressed–I was–that the faith’s national archives, located in the temple basement, include five full boxes of photographs, carefully labeled and sorted in chronological order, recording the temple’s progress from vacant land to final dedication. More valuable to the structural engineer are the architectural and structural drawings and even some preliminary sketches. Stecich unrolled all these in WJE’s Wacker Drive office the other day. The pile begins with frontal views and cross sections of the building, drawn by the architect in 1930; then come structural drawings, showing where steel girders, beams, and columns are embedded in concrete posts and beams. “If you don’t have any hint of where these are,” Stecich explains, “it’s hard to know where to start searching. I think of these drawings as a road map.” There is even a diagram detailing where the bent reinforcing bars are placed in slabs and girders, and another showing the crucial cornice gutter and the brackets and girders tied into the concrete underneath it. At the bottom is a set of drawings by John Earley–“center columns, 27 required,” “left columns, 45 required”–each with detailed dimensions. “This tells us which sections were precast, which were poured in place on the building.”

Stecich pauses, staring down at the wrinkled paper. “This is an awfully unusual drawing for an owner to have–a concrete shop drawing. I don’t think I can say how much easier this kind of information made the job. Number one, just having these drawings is great in itself–and number two, we were given them in an organized way. That might have happened one or two times out of 400 jobs I’ve done.”

All that information is welcome on such a complex building. Stecich describes the temple as having “four stories, the bottom two arranged as nine-sided polygons and the upper two as circles divided into nine sectors, with each story slightly smaller than the one below”–not unlike a wedding cake, except that all that lacy icing is solid concrete.

But even on quartz concrete, time takes its toll. The building is perceptibly less white now than when it was built, and up close one can see another sign of age. In sheltered parts of the exterior, the concrete surface with its quartz pebbles remains almost as smooth as Earley’s craftsmen left it. But where the concrete faces the weather, the quartz pebbles stick out: the cement-and-sand paste has worn away, leaving the aggregate more exposed than John Earley intended.

Of course, this process is slowly coarsening and blurring the detail all over the building. Bob Armbruster is thinking about that, even though it’s not on his immediate agenda. “We have to decide at some time to make some kind of record of the form of these designs,” so that they can be faithfully replaced or restored. “There may be some way to record their three dimensions without making a physical mold–possibly through photography or some other technique.”

In his search for more pressing problems, Stecich found the worst in the decorative concrete at the top of the third “story” (the clerestory)–cracks, decay, and efflorescence between the bottom of the dome and the top of the topmost windows. And the probable cause was not far to seek.

When it rains, water runs off the dome skylight (recessed just inside the concrete shell) and down into a big copper gutter that encircles the building, nestled in the structural concrete and hidden from view below by the flowery decorations at the top of the clerestory. In the course of half a century of weather (before being replaced in 1984), the copper developed holes. Water began to seep down behind the ornamental layer, freezing, thawing, freezing, thawing. “Concrete is porous,” WJE’s William Coney reminds us. “Freeze-thaw action gradually reduces concrete to sand and gravel.”

Clearly this problem was going to require some immediate action–and some kinds of work with which WJE would need help or which they could not do. The top of the clerestory is 20 feet above the roof of the second story, so Stecich would need scaffolding for a close-up look. In order to verify his plausible conjecture about water damage, he would need to take some samples of the concrete back to the petrographers in WJE’s Northbrook laboratories–and that would mean drilling (as it turned out) 16 holes in the stuff, some as much as 6 inches in diameter. Then the holes would have to be repaired. And once Stecich made his final report, the overall damage to the building would have to be repaired as well.

“We could have hired almost any contractor to put up scaffolding and drill holes,” says Stecich. “But you need someone special for these repairs.” NSA Properties selected William Hach and Associates, a Bensenville firm specializing in concrete restoration. “Bob Joyce is the superintendent there, and he’s a super guy to work with, very cooperative. That is something we as investigators need. You can’t lay out a rigid plan and say we’re going to do A, B, C, D, E. You have to adjust the plan as you go along, depending on what you find out.”

Of course the clerestory level, like everything else on the outside of the temple, comes in nines, but Stecich figured there was no need for nine scaffolds at this point. Using binoculars and cameras with telephoto lenses, the investigators surveyed the suspect area from the roof below, and picked the most deteriorated ninth (it faces south), the least deteriorated, and an in-between section. Those three got pipe scaffolds, from which the engineer made sure that appearances at 20 feet had not been deceptive. Then the workers started on the essential work of cutting cores (doing their best to avoid hitting the steel reinforcing bars). Without these little cylinders of concrete, Stecich couldn’t confirm his diagnosis, and Joyce would have less information on which to base his repairs.

“When you’re investigating a building failure,” says Stecich, “it’s not enough to say, ‘It did (or didn’t) meet code.’ You want to know how the structure actually performed.” Likewise in this case, where no serious failure has happened yet, it’s not enough to settle on the most plausible villain–the leaky gutter–without making sure. The truth might be worse: the steel reinforcing bars could be corroding, a genuine disaster since (in Coney’s words) “rust . . . occupies significantly more space than the original metal.” A corroded reinforcing bar is thus a long-fused bomb inside the concrete. (Fortunately the temple reinforcements proved sound.)

Another possibility is that the concrete might have been poor quality to start with. From the documents, this does not seem likely. According to Earley biographer Frederick Cron, “At a time when structural concrete was commonly designed for a compressive strength of 3000 pounds per square inch, the Earley Studio was routinely making concrete castings with a compressive strength of 5000 pounds per square inch.” And WJE petrographers’ analysis of the cores themselves bore out the paperwork: Earley’s high-strength specifications seem to have been followed very closely.

In the end, the culprit was what Stecich had surmised–water damage. The clincher was positive evidence under the microscope. As WJE petrographer and principal Bernard Erlin explains it, “When something freezes, it freezes from the surface inward, not all at once. So there will be a kind of thermal front moving through the concrete from front to back. So we look for fractures or fine cracks oriented parallel to that front, parallel to the surface.”

OK so far. But if water was indeed the culprit, how did it get in there, what other damage had it done, and how could it be kept out in the future? Stecich couldn’t tell enough by peering into inch-and-a-half diameter holes, and soon he had the contractor taking a concrete saw to the south-facing ninth of the temple and cutting out a two-foot-square access panel–“to be sure we know how the building was made,” and to a small extent, unmade.

Of course the panel didn’t want to come out even when all four cuts had been made; workers had to take a second slice along the top and another along the side before the 200-pound hunk would budge. (It now lies inside the temple at the clerestory level, where I took an experimental and fruitless heave at one corner. “How did you get this down off that scaffold?” I asked Keith Niles, who works under Bob Joyce. He gave a big grin: “Very, very carefully.” He is now contemplating how to get it the rest of the way down to ground level.)

“There!” sighs Stecich, showing me a photograph of the wall with access panel removed. “All of a sudden, don’t you get a much better feeling of what’s going on with the building?” I do. The reinforcing wires–visible in cross section–are OK, and so are the anchors that hold the decorative concrete to the undecorative concrete structure beneath. But there is a one-inch gap in places between the two concrete sections–“a place for water to be,” a gap that does not exist on the good section. “That,” says Stecich, “starts explaining things. The amount of efflorescence and the evidence of freeze-thaw action indicated a lot of water.” Joyce says the old gutter must have had a pretty substantial hole, because they found straw and feathers washed into the gap as well.

The structural concrete is substantially all right, and Stecich is pleased to see that the water found a path that mostly bypassed it. As a final experiment, they plugged the gutter and filled it to the brim. Sure enough, water came through pinholes in its side and showed up promptly in the access panel Stecich had cut out.

Every home owner with a pesky leak should have this kind of time–and patience. “When you isolate the source, seal it up and the leak stops, then reopen it and the leak starts again, you’ve got it,” says Stecich. As long as the copper gutter above and behind it is in good shape, the damaged concrete can be repaired or replaced without any fear that it will suffer the same fate.

Repair? Replace? With what? Good-quality quartz (not shot through with brownish-orange hematite) is hard to come by these days. There’s reason to believe that even Earley increasingly had trouble finding a supply as the job went along. “I talked to a guy,” says Joyce, “who said he knew right where Earley’s vein of quartz was. It’s a subdivision now.” He chuckles. “We were getting desperate for a while. We even thought of buying a house there and making a hole in the basement.”

Apparently Earley had his quartz shipped intact. Then, says Joyce, “they high-graded it. They’d go through and chip out all the brown stuff,” and then Earley would use his own crushers to make pebbles and sand to his specifications. After some time, Joyce found a supplier in Colorado who will let Joyce’s people pick the best quartz and who will sell a mere 40,000 pounds of aggregate–“that’s a semiload, the minimum amount he’ll deal with us for,” says Joyce. The usual minimum is 200,000 pounds.

This work (beyond plugging the core holes and the hole where the access panel was removed) won’t start until next year. Meanwhile, Joyce’s crew has cast 28 different samples in an attempt to produce a mix that will match the temple walls in color, size of aggregate, and ratio of aggregate to cement. To help them match, Armbruster has had small sections of the wall cleaned, so that the new sections will match after the restoration is complete and the entire building is cleaned. (It can’t be cleaned first, because the restoration process will itself raise dust that will darken it again.) The first test pieces they cast were a foot square; even the six-inch-square pieces are noticeably hefty.

Joyce isn’t sure he can rely on the petrographers’ core analyses in his work: “Different sectors [of the temple] look different. And we’re concerned about matching the outward appearance,” not the precise physical makeup of a sample core, most of which would be invisible to any outside observer. He has found that he gets a good match with about 20 percent fewer quartz pebbles than Earley’s specs called for. How come? “Well, we talked with Charles Meyer,” who worked in Earley’s studio and is now semiretired in Virginia. “He said they were still arguing with the engineers every day [on the job]. They would alter the design mix as they poured. He led me to believe they were making a lot of on-the-job changes.”

One change Stecich plans is for the new concrete to be air-entrained, a technology not available to John Earley. To resist future freezing and thawing, this process scatters air bubbles throughout the concrete to act as tiny expansion joints, so the material can give a little without cracking.

“Today,” reflects Joyce, “people wouldn’t consider putting up an exposed-aggregate building without waterproofing. They’d say it wouldn’t last.” Why has this one lasted, then? “Good workmanship,” he replies promptly. “Tight joints and square. And a rich mix–a lot of cement [and a minimum of water]. And the quartz aggregate.” He doesn’t say the words, but the thought hangs in the air: they don’t make ’em like this anymore.

After my stint up on the scaffold, Joyce and Niles head down to compare some samples they have made to match the stairs leading up to the temple entrance. I go with Armbruster for a quick look at the new 1,200-pane glass skylight under the dome. When he and I get downstairs, the others are pondering one more of the small mysteries that give the job its charm: Why are the quartz pebbles on the stairs all rounded?

First answer: people walking on them, about 200,000 a year. But that can’t be it. The pebbles on the vertical sections around the steps are as rounded as those on the treads.

Is it weathering, then? That can’t be it either, because then the quartz pebbles in the clerestory walls would be rounded too.

Finally Niles is inspired. “I’ll bet this was sandblasted”–and, of course, it was. “Look at them close, and you can see the etching on them.” Unlike the clerestory quartz, which is fractured clean, smooth, and shiny, these stones have been pitted as well as rounded. “We’ll be able to tell for sure,” says Joyce, “when we see them break the cores” taken from the stairs. But we don’t have to wait that long, even. A few steps below, one of the stair treads has recently lost its surface. The original quartz is exposed, and sure enough, it is not round but just as sharp and shiny and full of light as when it emerged from John Earley’s rock crusher almost half a century ago.

Art accompanying story in printed newspaper (not available in this archive): photos/Mike Tappin, National Baha’i Archives.