The Window-Light Workers
Flat Glass Production

Rev. 2006-04-19, 2008-07-05, 2009-04-12, -04-13, 2010-02-24, 2011-01-26, -02-03
Back to Good Old Days Center

Link to

Glass History Central Connection
Broad Glass
Crown Glass
Cylinder Glass
Old Glass Photos
COMMENT - This page includes two long quotes from the book "Glass Gaffers of New Jersey" about the men who blew crown and cylinder glass in the 19th century.  In addition, this page offers information on the 6 major methods of producing flat window glass down through the years referenced from other pages.
Cast Broad Crown Cylinder Drawn Float
Cast or Plate
Glass is poured on a slab and a heavy roller is worked over it, running on side rail spacers.  Diderot has many pictures of the process including special rectangular "ladles" in two sizes for a half and full sheet. These have slots cast in the side to permit carrying by men with rods like a stretcher or with a forked wheeled device and lifting in a rig that permits dumping.   The English text says the slab is copper and the side rails are removed when the glass is firm enough to not spread but before they can stick and the whole large sheet goes into the annealer for up to 10 days, which seems high to me. The sheet is quite large, perhaps 3 or 4 feet by 6 from the figures in the pictures. The glass that comes out is almost opaque from the scale of the copper and the rough rolling and must be ground and polished. After much work it is suitable for mirrors.  A large capital investment is required in equipment, furnace, and especially multiple annealing ovens.
Modern stained glass with patterns in it is made in the same manner on a much smaller scale - one or more colors of glass are ladled onto an iron slab, stirred if a swirl pattern is desired, and rolled with a roller weighing a couple of hundred pounds.  The roller or the slab or both may have patterns or be smooth. 2010-02-24
Diderot image of broad glass sheet workshopBroad glass, sometimes mistakenly called cylinder glass, was created in the 13th and 14th centuries, to make much smaller sheets than the later cylinder glass (below).  A rough cylinder is blown and the end is opened and supported with an interior punty and broken free of the pipe.  The diameter of the cylinder is opened out and it (from the illustrations) is cut from both ends with large nippers.  The cut cylinder is forced open on a table and then annealed.  It must be ground and polished for use because of the wrinkles the unrolling introduces and the roughness of the table.  Broad glass produces much smaller sheets but requires much less capital investment, the cold sheets can be annealed on edge, several to an oven.  The English version of Diderot does not mention the several reheats that the method would take, but these would be quick with thin glass.  The editor says the sheets "must be fairly small - about three feet by four" but the size on the table looks more like 2x3 feet and the tool image (Plate 252) he is referencing, which has a scale at the bottom, suggests that 40 inches by 20 is more likely. Scaling from the graphic on screen with pixels suggests 3.94x3.36 ft is possible from 1.25' diameter. 2010-02-25
Crown glass is spun to a disk with a boss in the middle which means it is cut to panes of moderate size - 12 by 8" would be exceptional.  As Diderot points out, the technique requires greater skill and handling heavier amounts of glass. To get nearly uniform thickness requires careful shaping of the vase shape that is spun out.  But the resulting glass is untouched by any surface so is polished and clear without further work.  It is also chilled when done and thus can be annealed on edge although it is shown being inserted in a slot just big enough to take one flat. (Plate 246) More from Glass Gaffers 2010-02-25
Cylinder glass was developed in the 19th Century and involves blowing an large long cylinder from a platform rig where the gaffer leans over to suspend the glass vertically with some risk of falling as described below.  The end cap of the cylinder is cut off, then the pipe end removed with the cylinder laying down and the glass is annealed so people are shown carrying them around on their shoulder.  The cylinder is cut lengthwise and laid in a sagging kiln where it is maneuvered as it opens out to a flat sheet.  Since the glass is much cooler and is not forced, it is in much better condition than in the broad method, while still producing a large rectangle.
Toward the end of the century, a method was developed of mechanically pulling the cylinder from the molten metal while maintaining pressure to produce a much longer and larger diameter cylinder than a man could handle. But the same technology led to the next type. 2010-02-25
This set of images from a video shows modern cylinder glass being made on a much smaller scale.  In the Lamberts video, the glass is shown being scored cold and flattened after reheating, unlike broad which was cut and flattened while first hot, although similar shapes and techniques are shown in Diderot's glass images for broad glass. 2011-02-03
When it was realized that a wide thin sheet of glass could also be drawn like the cylinder, at first a sheet was pulled and then cut off.  But a next step was to bend the glass at the top of the pull over a roller and into the annealing lehr [long oven with gradually reducing temperature] so that glass could be produced continuously slicing off the ribbon in pieces as it cooled.  The curving path introduced ripples that were noticeable but the finish was good and the automated continuous operation made large glass panes cheaper. Windows such as those in my 1926 house show the ripple.  Almost all window glass from 1900 to 1960 was made this way and the manual handling of glass mostly wiped out. 2010-02-25
In the mid-1950's, Pilkington in England developed a method of flowing glass onto a layer of molten tin under an atmosphere of nitrogen to keep from oxidizing the tin with gradual cooling,  The resulting glass has a totally ripple free surface that is as good as the best polish and ground plate glass at much less expense.  The company licensed the method as well as making the glass themselves and since about 1960 all window glass in the world is made with the float method. 2010-02-25

to the two types of glass discussed in these quotes, long before this time glass was made by pouring glass on a flat surface and rolling it with a heavy stone or iron roller - cast glass - and by blowing a cylinder and cutting it open while hot - instead of cooling it and cutting the ends off and then flattening it at kiln temps- called broad glass.  Both of these produced roughly textured glass.  Cast glass had to be ground to make clear plate glass.

from p.62 The Glass Gaffers of New Jersey, copyright

"A CASUAL "ANTIQUER" seeing bull's-eye window panes, with their center swirl of dense green glass, can't be blamed for assuming that these decorative bits were intentionally made that way. Actually, bull's eye panes are castoffs put to good use by thrifty pioneer settlers.


"The first American window glass, a great advance over hides or sheets of mica at the windows, was nevertheless most primitively made, by the crown method. The blower would form a pear-shaped globule of glass which he then moved down to the "bottoming" hole, in front of which was a low wall to protect the man against the heat, where he twirled the parison which caused it to spread out. A helper then stepped up with a small gather of glass on a punty rod that he attached to the opposite side of the "blow." `Then another workman touched a cold iron dipped in water to the blowpipe side, causing the pipe to separate. With one end of the parison now open, more twirling before the fire made the object basket-shaped.
"Suddenly the blower would twirl the punty rod so fast that the basket, with a loud ruffling noise like the snap of a flag in the breeze, would flatten into a smooth round plate. This process, described in an 1837 pamphlet, "Conversations on the Art of Glass-Blowing," printed by Mahlon Day at 374 Pearl Street, Manhattan, was called flashing. The disk, about 36 inches in diameter, was moved to a flat surface, separated from the rod and placed in annealing ovens, warmed only to red heat. Sometimes the cooling process took two to three weeks.
"Small square or diamond panes were cut from the circle of glass, to be fitted into mullioned frames. The thick knot where the rod had been detached was excellent for panes around doors as the glass was translucent but not transparent. The legend that the bull's-eye panes were installed in order to deflect Indian arrows is-well, just legend.
"There is substance for believing, though, that bull's-eyes in Jersey and Delaware houses built from 1740 to 1780 were made at Wistarberg, for there was no other window-light place operating then in those rather isolated colonies and Wistarberg certainly was using the crown method, as the cylinder process was a 19th century development.
"The Hon. John T. Bodine who supplied most of the data on Jersey for Weeks' 1884 U.S. census Report on the Manufacture of Glass stated he believed that Columbia on the Delaware which operated from 1812 to about 1838 was a crown glass works. This seems unlikely as the glasshouse depicted in the Thomas Birch painting is two stories high, needed for cylinder glassmaking but unnecessary for the early bottle works or crown glass blowing. Some sizes of glass advertised for Columbia, moreover, were much larger than could usually be cut from a disk of crown glass."

Old time glass blowing factory

Fig. 45 WINDOWLIGHT GNOMES in fiery Pittsburgh scene of 1876. Note cowl hoards used by five workmen; trimmed cylinders on cradle in foreground. From Every Saturday, March, 1967. Courtesy of Corning Museum of Glass.

* Conversations on the Art of Glassblowing, printed in 1837 in New York City, is written for the laity, so the cylinder method described must have been well established by then. p.62 The Glass Gaffers of New Jersey

"Window-glass made from handblown cylinders or rollers was a transformation of the early 19th century, one that opened the way for large panes and clearer, eventually nearly flawless, glass. Cylinder glass was as revolutionary for its time as the continuous-tank method and machine-made window-glass which were in the agonies of creation at the close of the 19th century.
"Some of the greatest showmen and unsung heroes of the glass industry were the men who blew, entirely without machines, window panes by the cylinder method prevalent from the 1830s to the early 1900s.*
"Their story, rarely told, comes alive in the words of Charles Westcott, Jerseyan and 76-year-old blower of precision laboratory glass, whose father, grandfather and great-grandfather were window-light blowers. (Old-timers never speak of windowpane blowers.) The sojourns of these four generations show, like the travels of Samuel Huffsey, how far afield Jersey glassblowers ranged, and accordingly how difficult it is to affix the provenance of many collectors' wares made in the Jersey tradition.
"Jesse Wescoat, Sr., grandfather of Charles Westcott who changed the spelling of the family name, began blowing cylinders of glass at the Batsto Works established in September 1846 by the Richards family when they saw the fortunes of their iron empire toppling.
"An iron blowpipe and two iron blocks or molds for shaping window-glass parisons, now owned by Charles Westcott and on exhibit in the Batsto Museum, were the tools of Jesse Wescoat, Sr., and his descendants. When the glass project ended in 1867 and Batsto became a ghost town, the senior Wescoat packed up his tools and went to blow at nearby Waterford, noted for good workmanship, then later at Winslow, also of high repute, on the same stage road.
"These tools descended to Jesse Wescoat, Jr., who as a window-glass blower used them at Glassboro while working there. Daniel Wescoat, born at Waterford in 1859 in a white house still standing on the main road, and father of Charles, also used the tools at Glassboro. Charlie recalls as the happiest days of his life his early childhood in Magnolia where his father in 1901 was a windowpane blower for the Magnolia Glass Company, an outfit unrecorded in any standard works on glass.
"From Magnolia, Daniel moved to Shingle House in northwest Pennsylvania to blow for the window-light furnace in this little town. Finding blowing too strenuous he became a flattener. When a union dispute about 1907 caused a work stoppage, Daniel left his tools on the premises in expectation of being called back shortly. Blowing did not resume and Daniel moved his family back to Vineland. Years later Charles retrieved the blocks and pipes from a blower at Shingle House."
"'When I went to look for the tools no one even knew where the window-light factory had been but I recognized it right away. They're usually 2 -story frame buildings,' Westcott said. The Mahlon Day pamphlet describes the furnaces of crown works as conical. Those of cylinder works were generally bottle-shaped, with a powerful draft. Even a bottle-glass furnace needed a fearful draft. Ernest C. Stanmire, Jersey glassworker and antiques dealer, recalls seeing a man at Moore Brothers in Clayton sit down to eat a sandwich, only to have it whisked irretrievably up the chimney.
"It took a certain constitution and co-ordination to be a proficient window-light blower, and not many men could qualify.
'Those old window-light blowers were the proudest people in the world,' Charles Westcott recalls. 'Everybody realized how highly skilled they were, and the blowers took exceptional pride in their talent. They had to be real men. They combined fast thinking with brawn.'
"Brawn the trade certainly demanded. For the iron blowpipe hefts about 30 pounds, and the glass gather added about another 50 or 70 pounds. All told, the blower not only had to lift at least 80 pounds of pipe and lethally hot molten glass but also had to manipulate all swiftly, adroitly, so as to avoid breaking the glass and burning himself and his fellow workers. A split-second lag in reaction could mean disaster for the whole shop.
"Window-glass blowers usually worked at a 4-pot furnace, two men to a pot. At peak operation, white-hot fires blazing and a dozen men and small boys rushing about in dangerous proximity, the scene was a confused inferno.
"The blower, wearing boots and stripped to the waist, ascended a ramp or steps to a platform where he could work on a level with furnace doors. Below him was a pit. On a string around his neck, the blower usually wore a cowl board, a face-shaped wooden mask, eye-slits fitted with amber or blue glass. Ready to approach the searing flames, the blower clamped the mouthpiece of the cowl board between his teeth, his sole protection against the murderous heat.
"His smaller or dipper-shaped iron block, though used first, was called the 'second glass block.'  The gatherer, not the blower, picked up a blob of liquid glass on the end of the pipe, turned it in the smaller block to give some slight form, then kept adding more metal in about three or four trips to the furnace until he had about 60 pounds of white-hot glass which he had in the meantime expanded to rude shape.
"The blower now began blowing this parison into an iron block of about 18 inches diameter, until the parison seemed to shape to a similar diameter. The "blow block" had meantime been placed a-tilt in a wooden tub or barrel cut down to a height comfortable for the blower. Sawdust was usually sprinkled in the block to keep the glass from adhering to the iron, and the block was set in water as a coolant. By the end of this process the water was usually boiling.
"Now with the block as a guide the blower blew the glass balloon until it was several feet long. Then he took this fragile roller to the swing-hole, a drop-off of at least 9 feet below the platform or foot-bench where the worker stood. While swinging the blowpipe like a pendulum, the crouched blower stretched the glass cocoon to the wanted length, about six feet, sometimes more.
"Pull of the iron and weight of the glass were so great that many of the men wore a chain, one end attached to their belts, the other to a post, so that they would not be dragged head first into the swing-hole filled with the broken glass of rejects. Blowers who scorned to wear a chain were often pulled forward and crashed into the hot metal and broken glass in the pit below. Some never came out alive.
"Window-light blowers wore heavy stripping around their wrists to give them steel-like control of the great weight. The longer they manipulated the parison "the heavier it got." The blower could not relax for an instant before the roller was deposited in a wooden cradle on a sawhorse to cool.
"The critical point, however, came before that stage. It was the opening of the far end of the roller, in this ingenious manner. When the roller was the proper shape and length,

The Glass Gaffers of New Jersey p.64

the blower gave it a mighty heave up to an iron crane in front of the furnace so as to reheat the cylinder. With both his thumb and the blowpipe in his mouth, the worker blew an extra puff that trapped air inside, as he immediately capped the pipe with his thumb. The expanding hot air forced open the far end of the roller. It was now ready to transport to the cradle. Some of the showmen-and there was often a cluster of visitors-would swing the roller and blowpipe over their heads in a great virtuoso display of strength and skill. But sometimes things went awry and the hot glass broke into myriads of fragments that showered down on the workers.
"When a roller had been carefully deposited on the cradle a helper touched the glass near the blowpipe-end with a wet iron and this magically released the cylinder from the pipe. The roller was now a long pale aquamarine pod with rounded ends. After the glass cooled a bit these ends were cracked off. To do this a thin strip of molten glass was laid on so as to encircle the cylinder where separation was wanted. Again the ends were touched with wet metal and the tops fell off, leaving an open-end glass cylinder. Usually retrieved, the caps were prized as bells under which to display wax fruit.
"Next, a workman threw a handful of sawdust the length of the cylinder, the curved surface of which caused the sawdust to fall to the bottom. Then a hot iron rod was run back and forth along the bottom of the roller until it was grooved. Once more, the glass was touched with a wet iron for an instant and, lo, the cylinder split neatly down its entire length. Now a curved sheet, it was carried to the flattening ovens where renewed heat made it flexible, so that it could be leveled to a plane surface. In the final step before cutting into panes, the large sheets were placed in a "dip," a barrel-like container full of boiling water, to be tempered for a half hour or so.
"Window-light blowing was so well paid in relation to other glassblowing that a man couldn't get into the trade unless he was the son or the brother of one of these vastly proficient craftsmen who, the theory was, would teach his relatives the secrets. No doubt about it, window-light blowers had status.
"Yet it was the cutters who were the only glassmen who came to work wearing a collar and tie. Their skill was so essential to high production that they were paid even better than the blowers.
"The cutter, with both arms outspread to hold the large oblong of glass, would stand the pane upright on the cutting table. Then, to the consternation of small boys and other uninitiated souls, the cutter would let the sheet of glass fall to the table, which it did without breaking. An air cushion protected it-usually. Cutting into small panes was done in the 1850s or so with a pencil-shaped tool that was diamond-tipped. Harder than glass, the diamond was simply drawn alongside a ruler and neatly cut out panes, provided the sheet did not have too many "stones" and air bubbles. 'The job looked as easy as separating a soda cracker from its mate,' said Westcott.
"Because window-light work was so strenuous, union rules in the early 1900s provided that blowers were not allowed to make over eight cylinders an hour. A day's work was 64 cylinders, but many men could not turn out that number. About 1902 a window-light blower could earn $90 a week, but the average was more like $60, still high for the times when unskilled labor was paid about a dollar a day."


Many forms of clear glass are still being made that either use the methods of old glass making or make variations with newer cheaper methods.  One result is glass like the images immediately below, unfortunately for my pursuit of authentic old glass.  Some of the many variations sold today can be found at this Delphi Glass site Clicking on the small images will show a larger image and a brief discussion of the relation to older glass.  This link provides a central listing that includes Spectrum's explanation of the relation of their clear glass versions. 2011-01-26
These four pictures were taken at the Washington's Headquarters in the Ford mansion at the Morristown National Park, a winter encampment of the army in 1777, because I liked the shape of the window and Texture in glass, modern reproduction of 1700's glass Old window in Washington's headquarters
Old glass reproduction in Washington headquarters mansionBlown (or rolled) distortions in the glass.  I have no idea whether this glass is truly as old as the house and I failed to ask the guide, but it looked neat and the pictures came out ok. But ...
"Thank you for you interest in trying to have information on your web site as accurate as possible. The oldest glass in the Ford Mansion is from the 1930s. The glass that has the deep swirls is from the 1980s. It is an attempt to reproduce glass that looks like it is from the Revolutionary War era, but now we do not think it is a good attempt.
I hope you enjoyed your visit to Morristown National Historical Park and return again sometime in the future."
Anne DeGraaf, Chief of Interpretation, Morristown National Historical Park,
Old glass reproduction in Washington headquarters mansion
Window glass, drawn, ca 1926 probably If you take my word that the fence, window and venetian blinds in the image at left are perfectly normal then you get some idea of the detailed distortion of window glass that was drawn up and then pulled over a roller to be annealed in the 1920's.  The degree of distortion is enhanced by using a zoom lens to view through a moderate portion of the glass (i.e. it doesn't look this bad to the eye, especially when closer to the glass.)
 This window is in the bedroom of my house built in the late 20's east of downtown Dallas and is almost certainly the original glass.  The distortion of the evenly spaced mini-blinds, the window frame and mullions and the even fence top are caused by slight horizontal ripples in the glass.  The camera was moved back 10 feet or so from the glass and then zoomed to view through about 8" square on the glass which is a large pane (about 28" square) unlike the small panes in the window opposite. 2008-07-05 [Click the image for a much larger view]
I would welcome additional images of clear glass that show distortion typical of a type of glass, especially when the glass can be exactly dated.  It is difficult finding good examples of glass that has not been discolored by being in the ground and is really old. It is easiest to take pictures of the distortions themselves when the background is small and uniform (like leaves) and highlights are reflected in the glass.