Incandescent Lamp Carbon "Filaments"
In the early volumes of the Western Electrician Philip Atkinson published a series of articles under the title "The Elements of Electric Lighting." The articles covered a wide range of topics and included among them were discussions of how some of the lamp manufacturers made their carbon "filament" light elements. These comparisons were valid for the year 1888. It is of interest to take advantage of Atkinson's fine work to put these different methods down in a more current information source.

The method first tried by Thomas Edison was mentioned before his successful use of bamboo was detailed. Regarding bamboo Atkinson said1:

The Edison Carbons
"In 1880 he (Edison) patented the process which, with some modifications, he still adheres to. In this process he uses filaments of bamboo, which are taken from the interior, fibrous portion of the plant. The cane, after being cut into sections of the required length, and the hard outer surface removed, is split and shaved down into flat strips, which are then pressed between dies, and fine filaments of the required length and diameter obtained. These are placed in molds made of nickel, in grooves of the required horse shoe form, and closed so as to exclude the air. The molds are then placed in muffles, and the filaments carbonized at a very high temperature. They are then attached to their platinum wire supports, by electro-copper-plating, and introduced into the little lamp globes. The lamp is then attached to a Sprengel air pump, and during the process of exhaustion the filament is alternately heated and cooled by an electric current; the temperature being gradually raised by increase of current after each cooling, till a high degree of incandescence is attained. This removes all the occluded gases which remain after the carbonizing, and renders the carbon homogeneous, elastic, and refractory at a high degree of temperature. As this is a far more severe test than any to which the carbon will be subjected in use, all imperfect and defective carbons are destroyed in the process, and only the best survive."
Atkinson continued to outline the procedures used by Lane-Fox2, Cruto, Swan, Weston and Bernstein3.
The Lane-Fox Carbons
"In the manufacture of the Lane-Fox carbons the raw material is obtained from the bass broom, a species of grass fibre. The hard outer surface is removed by immersion in a solution of hot caustic soda or potash and subsequent scraping, after which the alkali is removed with boiling water, and the fibers, in lots of about 100, bound to blocks of plumbago shaped so as to give them the horseshoe form. These blocks, in lots of 50, are imbedded in powdered charcoal in plumbago crucibles, placed in a furnace and subjected for twenty minutes to a white heat. The fibers after being carbonized in this manner are gauged and sorted, and all of the same diameter placed together, after which they are suspended separately from spring clips by attachment at the ends, in large globes filled with a hydro-carbon gas obtained from benzole or coal, and subjected to the process termed flashing, first proposed by Sawyer. This process consists in rendering the filaments incandescent, either by the passage of an electric current or by the heat of a furnace, the latter being the method adopted in the Lane-Fox process. The gas in contact with each filament being decomposed, a layer of hard carbon is deposited which renders the filament denser, smoother, more homogeneous, more durable and more uniform in size; the smaller parts becoming hotter than the larger and thus acquiring a thicker deposit.The increased diameter reduces the resistance and the process is continued till the resistance required for lamps of a given candle power, ranging from 16 to 60, is obtained."
The Cruto Carbons
"These are made by using as a base a fine platinum wire bent into the horse shoe form upon which the carbon is deposited by the flashing process. This is effected by attaching the wires, placed in a long glass vessel, to insulated metallic supports, by which an electric current can be sent through each of them. A current of olefiant gas C2H4, circulates through this vessel, the gas being made from alcohol and sulphuric acid, thus furnishing a gas of pure hydro-carbon. The platinum wires are connected with a shunt circuit so that an electric current passing through them can be graduated to any required strength by a resistance varying from one ohm to 200 ohms. The wire being heated to incandescence in this manner the gas is decomposed, and carbon deposited. To insure a uniform deposit it is necessary to guard against the influence of the earth's magnetism by placing the wires in a plane at right angles to that indicated by the dip of the magnetic needle; and in the latter part of the process the current is reversed. This operation, which requires about two and a half hours, produces a filament remarkedly compact and homogeneous, and uniform in cross section and resistance. Special care is required to maintain uniformity in the successive stages of the process, otherwise the result is a filament of dissimilar superimposed layers of carbon, lacking homogeneity, and practically worthless."
The Swan Carbons
"These carbons are made from cotton twine, prepared by immersion in sulphuric acid diluted with one third part water, by which they attain a consistence similar to that of parchment; they are then thoroughly washed to remove the acid, reduced to a uniform cross-section by being passed through disks, after which they are wound on rods of carbon or earthenware, each in the form of a flat spiral having one convolution, and carbonized by being imbedded in powered charcoal in a crucible raised to a white heat. They are subsequently coated with carbon by the flashing process, and like the Edison carbons, heated and cooled alternately by the electric current to remove the occluded gases, while inclosed in the globe during its exhaustion. Each filament when finished is five inches long and ,005 of an inch in diameter."
The Weston Carbons
"The raw material used for these carbons is cotton or linen cellulose, which by the action of nitric and sulphuric acids is converted into nitro-cellulose—gun-cotton—which is subsequently dissolved in a mixture of alcohol and ether and converted into collodion, and finally has its combustibility reduced by the action of ammonium hydro sulphide or other chemical agent producing a similar effect.

"This artificial product, which Weston calls 'Tamadine,' is an amber-colored, amorphous cellulose of great ductiliy, tenacity, and homogeneousness. It is rolled into thin sheets between steel roller, and filaments of the sizes required for lamps of different candle power are cut from the edge of the sheet, bent on shapes into the horse-shoe form, carbonized in the usual manner, and flashed in hydro-carbon gas by the electric process. The carbons when finished are highly elastic, and have a smooth brilliant surface like that of a steel watch spring."

The Bernstein Carbons
"These are made from silk ribbon of the finest texture, woven into a hollow cylinder of the required size, from which sections of the proper length are cut, which are immersed in a thick syrup of cane sugar till saturated, and subsequently in melted paraffine; they are then bent on shapes into the horse-shoe form, placed in graphite molds of the same shape and imbedded in graphite powder, and baked for 24 hours. The subsequent flashing process is the same as that for the Lane-Fox carbons.

"These are the only carbons made from an animal substance, and with a hollow structure; and the advantage claimed for this structure is increase of surface and consequently of radiating capacity with a given degree of resistance."

Atkinson continued with his description of lamp filaments by commenting on their form:
"The shape of the filament varies, the horse-shoe being the most common; the Swan filament, as already mentioned, is a flat spiral, and the same form is adopted for the Brush, while the Maxim resembles a capital M with the angles rounded off. The Sawyer-Man filament resembles a figure 8."
The Weston filament was made in the shape of a horse-shoe as well as in a corrugated fashion. Atkinson then described the method of attachment of the filament to the lead wire.

"The filament is in all cases made largest at the points of attachment, either in the form given previous to carbonizing, or by the subsequent deposition of carbon on the extremities or their insertion into small perforated carbon cylinders. This is necessary for the purpose of better attachment to the platinum wires, but more especially to reduce the resistance and hence the strain on the filament at the points where the current enters and leaves, and to increase the strength where liability to fracture is the greatest. The attachment of the filament to the platinum wires is accomplished by different methods; that of Edison , by copper plating,... The Lane-Fox filaments are cemented into short carbon cylinders, to which also the platinum wires are similarly attached, with a cement made of India ink and plumbago. The attachment of the Cruto filaments is effected by their insertion into tubes formed on the ends of the platinum wires, and the subsequent deposition of carbon by the process on these junctions. The Weston filament terminates in carbon cylinders, nickel plated, each about 3/16 of an inch in length and 1/32 of an inch in diameter, around which are wound in spirals the flattened extremities of the platinum wires. This method is also used with the Bernstein filaments, and is attributed to W. Siemens."
Weston lamps with the two types of filaments are shown below, to the left3. Two Swan bulb designs are shown to the right4.

References
1) Philip Atkinson, "The Elements of Lighting", No 30, Western Electrician, Vol 2, No 7, Feb 18, 1888, pg 79.
2) Philip Atkinson, "The Elements of Lighting", No 31, Western Electrician, Vol 2, No 9, Mar 3, 1888, pg 115.
3) Philip Atkinson, "The Elements of Lighting", No 32, Western Electrician, Vol 2, No 10, Mar 10, 1888, pg 124.
4) Philip Atkinson, "The Elements of Lighting", No 33, Western Electrician, Vol 2, No 11, Mar 17, 1888, pg 134.