Although color is one of the more obvious and relatively easy to describe attributes of
a historic bottle, it is unfortunately of limited utility in classifying a bottle as to age or
type. One of the better discussions on this is from The Parks
Canada Glass Glossary by Jones & Sullivan (1989), and is quoted below:
"Because colour is a
universal attribute of glass and is convenient for mending and
establishing minimal vessel counts, it has been latched onto by some
archaeologists as a classification device. Although classification by
colour is simple to do, the end result is of little value for the
following reasons: colour does not have a direct relation with glass type
(the common green, amber, and brown glass colours can occur in soda,
potash, and lime glasses; many lead glasses are coloured); colour is not
related to the technology of glass object production (i.e., it has nothing
to do with whether the glass is free blown, mould blown, pressed, or
machine made); colour is only weakly related to the function of the object
(almost all colours can be found in all types of objects, an obvious
exception being "black" glass which does not occur in tableware).
Given these factors there is little justification for using colour as a
means of classification. There is a very broad chronology of
popularity of various colours over time; however that chronology cannot be
applied to individual glass objects with any significant level of
meaning..."
Glass Color - How does
it occur?
The majority of common bottle
glass is "soda-lime glass" which is primarily composed of silica, soda
(aka soda-ash) or potash, and lime
- the latter two ingredients often referred to as the "alkalies" (Hunter
1950; Toulouse 1969; Munsey 1970).
The silica (silica dioxide) typically makes up 60-80+% of the glass
composition and is primarily derived from sand. The purer the sand
(i.e., the higher the silica concentration and less iron) the better, as it
is the other impurities - desired or undesired - that give glass its
color. Low iron means more control over the ultimate color (Hunter
1950; Tooley
1953).
Glass which is composed of pure silica (99.9%+) would be colorless glass.
However, making glass from pure silica is not practical or commercially
viable because of the prohibitive expense of acquiring such in its pure state
and the much higher temperatures needed to properly melt. Soda
(sodium dioxide) - aka "alkali," "soda-ash," or "potash" in the trade (Trowbridge 1870;
Toulouse 1969) - is
added to the sand as a "flux" to lower the melting temperature of the
silica. Lime (calcium oxide) is added to the batch as a stabilizer since simple
glass made from just sand and soda ("water glass") is water soluble making
it of little use when formed into a bottle (Tooley 1953; Kendrick 1968; Jones &
Sullivan 1989). Broken glass (aka "cullet") on hand from misblown,
broken or returned bottles was also often added (Toulouse 1969).
From
this point in the glass producing process, the final color of the glass is
a matter of both controlling off-coloring impurities and achieving the
desired color. This is done by adding certain types of compounds to
the glass batch in certain quantities. Bottles made from glass with
just the basic ingredients (sand, soda & lime) will usually be
different shades of green because of the iron impurities in the sand, though
other colors can also be attained depending on many factors. So
called "natural" colors are those that result "naturally" from the basic ingredients in a glass batch (McKearin & Wilson 1978). In
general, with lesser amounts of iron or less oxidation of that iron,
shades of bluish to greenish aqua are achieved. With higher amounts
of iron or higher oxidation of the iron, darker greens will usually
occur (Toulouse 1969a; Jones & Sullivan 1989)).
In
order to create other colors, the iron needs to be variably neutralized
and appropriate colorizing agents
or compounds added to achieve the desired color. For instance, cobalt oxide added in
proper quantities to a properly prepared glass batch results in a
distinctly intense blue as shown in the bottle to the left. In fact,
this color is known as "cobalt blue" in the glass manufacturing world
(Scholes 1952).
Glass composition formulas were
(and probably still are) closely held glassmaker secrets as the experience
of extensive trial and error experimentation in glass making was not
readily shared with others. Variations in glass color resulted from
a myriad of different causes including the strata of the sand source, the
mineral in the soil of the of the trees burned to produce "potash" (an
"flux" alternative to soda), and many others known and unknown
(Toulouse 1969a). Many colorizing compounds work in different ways
depending on whether the glass pot environment is oxidizing or reducing
(Tooley 1953; Kendrick 1968; Toulouse 1969).
However,
discussing the simple
addition of chemical additives makes any discussion of glass making and
glass coloring too
simplistic. Glass chemistry is a complex science that is beyond the
goals of this website and will not be pursued here. For one who wishes to pursue
this subject, Tooley's (editor & one of the chapter authors) 1953 book "Hand Book of Glass
Manufacture Volume 1 - A book of reference for the plant executive,
technologist and engineer" is recommended though possibly
hard to find.
In the following color descriptions,
the different coloring (and de-coloring) agents or compounds for the
different colors are briefly noted. This is just informational
because the actual chemistry is of little utility and
glass colors only contribute a little to the process of dating or typing
historic bottles. It is, however, part of the overall "story" of bottles covered
by this website.
Having quoted this, color is still an
important descriptive element for the recordation and classification of bottles.
Bottle colors also warrant coverage here simply because they are of
fascinating interest to
people. As implied in the quote above, there are some time related
trends in color that can be of utility for dating. For example, if one
has a colorless ("clear") bottle which was de-colorized
with selenium and/or arsenic which gives the thick parts of the glass a subtle
"straw" tint, it very likely dates no earlier than World War I
(1914-1918) and infrequent in bottles after the 1940s or early 1950s (Kendrick 1963;
Lockhart pers. comm. 2003; empirical observations). The specific
"diagnostic utility" of a given color is noted in the descriptions below.
There are also some colors which where
very rarely used for one type of bottle (i.e., cobalt blue for cylinder liquor
bottles is very uncommon though do exist) but quite common in others (e.g., cobalt blue for poison bottles or Civil
War/Antebellum era soda water bottles). Thus, some information can sometimes
be gleaned from knowing what color is or is not likely to occur in a given
category or class of bottles. This may be especially useful in the identification of bottle
fragments (Bottle Fragment Identification).
Color Naming
Simply put, people observe or interpret colors (or in Canada - colours) differently.
Even the same bottle to the same person can
vary widely in color depending on differing lighting situations - direct
and indirect sunlight out in the field, fluorescent lights in the office,
and incandescent lights at home. Adding to the confusion is the jumble
of terminology that is used to describe colors and the almost infinite
color variations. As noted on Greg Spurgeon's fruit jar oriented website
there is no "governing authority" on glass or bottle colors
(Spurgeon 2004).
(Note: Spurgeon's excellent fruit jar color information webpage is located at the
following URL:
http://www.hoosierjar.com/colorguide.html.) There always has been and
will continue to be confusion as to color nomenclature even though many
attempts have been made to try to standardize it. The collector world
is rife with unusual naming, like "strawberry puce with apricot overtones" being one example of
a lengthy color name
which is intended to help
clarify the exact color of bottle but can often end up causing more confusion than
clarity.
Color names are usually analogous
with something that people are familiar with in the natural world like
"forest green", "sapphire
blue", "amber", "olive green" and so forth (McKearin & Wilson 1978).
When describing colors, modifiers can and should be used to help narrow
down the specific color. Examples of common modifying terminology
(adjectives) is dark or deep (for the denser end of the color), medium
(mid range density), light (paler density), as well as clear (for
noticeably translucent glass of any color) or murky (glass with
diminished clarity). Dual color naming (e.g., olive amber, yellow amber) is
a very common and a useful way to describe colors that are shades of a
primary color category. For example, "yellow olive" is a dominantly
olive color with a shift towards yellow, whereas "olive yellow" would be a
dominantly yellow color with a slight, but noticeable, olive tint (Spurgeon
2004).
This author of this website does not have any
delusions of this being the "final" word on colors or color naming. It is considered useful,
however, to briefly describe, name, and picture some basic
bottle glass colors that are noted on this website. Though this page does
not cover every color possible, it does cover those colors that probably make up
95-99% of the bottles made during the era covered by this website - the 19th
through mid-20th centuries.
The information on this page is a composite taken
from numerous references which are noted where appropriate throughout the text.
The most important references were: Scholes (1952); Tooley (1953); Kendrick (1968);
Toulouse (1969); Munsey (1970); McKearin & Wilson (1978); Fike (1987 & 1998);
Jones & Sullivan (1989); U. of Utah (1992); Spurgeon (2004); and many other
references to a lesser
extent.
The following glass color description
categories are not in
any significant order except that the list moves from the generally lighter to
darker colors. A user can either click on the color specific links below
or just scroll down through the descriptions to find the color
that you are interested in or that matches the color of a bottle you are trying
to gain information on.
Each color category discussion includes a
general summary of how the color was produced, variations within the color
category, pictures of bottles (with dating details) showing some of the
variations, and an overview of the diagnostic utility of the color for
dating and/or typing bottles.
This
color is the actually the absence of any color. Colorless is
preferred over the term "clear" or "white"
glass since the former term refers
more accurately to the transparency of the glass not its color, i.e., "clear
green"; and the latter term implies milk glass which is discussed
below (McKearin & Wilson 1978). Colorless glass was a goal of glass
manufacturers for centuries and was difficult to produce because
it required the use of virtually impurity-free materials. Venetian
glass makers produced their crystallo as early as the 15th century
and glass makers in 18th century England
made what was known as "flint" glass from virtually pure quartz rock
(calcined flint) which
was simply called "flint" (Hunter 1950). Improved chemistry and glass making methods of the late 19th
and early 20th century allowed for process efficiencies that made
colorless glass easier and much cheaper to produce with the use of
various additives in the glass mixture. The term flint glass
was and still is used somewhat erroneously by glassmakers to describe colorless
glass that is made with low iron sand. It is, however, not true flint
glass. Flint glass is sometimes called lead glass (and
vice versa) though true lead glass is made with lead oxide (Dillon 1958; Toulouse 1969a;
McKearin & Wilson 1978). Colorless glass was also called "crown"
by early glassmakers (Hunter 1950).
Colorless glass is not
always, or even usually, absolutely colorless. It will usually have very faint tints of
pink or "amethystine" (faintly visible in the base of the bottle to the
left), amber or "straw", grayish green,
grayish blue, or grey. These faint colors are viewed easiest when
looking through the thickest portion of the bottle, i.e., sideways through the
base. Colorless glass is usually attained by using the purest sand source
possible and by adding "decolorizing agents" to the glass batch
to offset the residual iron impurities (Dillon 1958).
Common decolorizing agents were
manganese dioxide, selenium dioxide (usually in conjunction with cobalt
oxide), and arsenious (arsenic) oxide which
is also used as a stabilizer of selenium in decolorizing glass - or some
combination of these compounds (Trowbridge 1870; Scholes
1952; Tooley
1953; Lockhart 2006a).
Colorless glass actually does have
more utility in dating and typing than most other colors, though
still of limited application. Some of
the better dating reliability is for bottles with manganese dioxide
decolorized glass. Upon exposure to sunlight, this glass will turn a
light pink or lavender to moderately dark
amethystor purple depending on
the amount of manganese and amount of ultraviolet (UV) light.
This is called "sun-purpled" or "sun colored amethyst " (SCA)
glass. The
picture to the right shows a Johnson's Chill and Chill Tonic
(Savannah, GA.) manufactured ca. 1900-1915. This bottle began
its life as colorless glass and has "turned" a much darker than average
color of amethyst due to the exposure to (likely artificial) UV light
(Kendrick 1968; Lockhart 2006b). The light lavender tint produced by
manganese offsets the green tint of the iron impurities in sand creating a
largely colorless glass.
The Venetians
apparently discovered by the 15th century that manganese could be used to
decolorize glass. Manganese became known as "glassmakers soap" due
to the ability to "cleanse" or neutralize the effects of other impurities
in the sand, particularly iron (Hunter 1950). Manganese dioxide
induced colorless glass was, however, by far most commonly used from the 1880s to
about the end of World War 1. At that time manganese dioxide use was
greatly reduced for a variety of reasons, although largely because it did
not work as well as other chemical decolorizers (see next paragraph) in
the open continuous glass tanks used by the increasingly dominant bottle
making machines - both semi-automatic and automatic. It is often
noted in the literature that the reason for the switch from manganese
dioxide to other decolorants was due to the cut-off of German imports to the U. S. during
WW1. Although all imports from Germany (and Europe in general) were
greatly constricted during this time, Germany was not a significant
source of manganese for the U. S., providing only 2% of the imported
manganese supplies in 1910 just prior to the war (Kendrick 1964; Lockhart
2006b). It should be noted that occasional manganese dioxide
decolorized
bottles may date as early as the 1820s or as late as the 1930s (McKearin & Wilson 1978;
Giarde 1989; Jones & Sullivan 1989; Lockhart
2006a & b), although the
large majority of bottles made with manganese decolorized glass were made
between about 1890 and 1920 (empirical observations).
Colorless glass which was de-colorized with selenium or arsenic (or
typically a combination of the two in conjunction with cobalt oxide) results in a very faint "straw"
or amber tint to the thickest portions of the glass (Scholes 1952; Tooley 1953; Lockhart
2006b). The picture to the left shows this color evident
in the thick portion of a milk bottle (underneath the line pointing out
the valve mark) that dates from between 1925 and 1930 based on a makers mark for the Pacific Coast
Glass Company (Toulouse 1971). Click
Cloverdale Dairy Co.
to see the entire bottle which was used by a dairy in eastern Nevada.
This colorless "color" can be very diagnostic of a machine-made
bottle made from about 1912 to 1915 to typically no later than the 1950s (Girade
1989; Lockhart 2006b; empirical observations). The straw tinted
colorless glass in bottles does show up frequently in mouth-blown
bottles but typically later ones (1900-1920), although can be found
occasionally in
bottles from the mid-19th century. (Click
French mustard bottle to view an 1870s era bottle with a faint straw
cast - evident at the heel - to the otherwise colorless glass.) Selenium was the best
decolorizer for glass made in open glass tanks (versus the earlier closed pots) which was used with most all automatic
bottle machines. Like the colorless manganese dioxide glass,
selenium decolorized glass will react slightly to UV light which produces
or enhances the straw tint (Scholes 1952; Lockhart 2006a & b).
Diagnostic Utility: Both of the above
colorless glass tints can be
useful diagnostic tools for an archaeologist who may be dealing with
fragmental bottles. One can be quite confident that if the fragment
is colorless with a slight straw tint, it very likely is from a machine-made
bottle, unlikely to date from much prior to World War 1 (i.e.,
mid-1910s), and could date as
late as the mid-20th century (or later).
Conversely, a colorless fragment with a slight amethyst tint is quite likely
to date to or prior to World War 1 (1915-1920) and is more likely than not to be
from a mouth-blown bottle. Bottles with a grayish tint seem to
date between 1915 and 1925, although numerous examples outside that range
have been noted by the author (Giarde 1989; empirical observations). Generally speaking, bottles of
colorless glass were relatively uncommon prior to the 1870s but
became quite common after the wide spread use of automatic bottle
machines in the mid to late 1910s (Kendrick 1968; Toulouse 1969a; Fike
1987; U. of U. 1992). Nothing is absolute in these date range estimates, but
they are believed to have reasonably high reliability.
As a side note,
crown top soda bottles were generally
not decolorized with manganese after 1914, giving a good ending date for
such "colorless" bottles with an amethyst (manganese dioxide decolorized)
cast to the glass as the beginning of World War 1; most of these type
bottles would be mouth-blown (Lockhart 2006a & b).
This
color - like most of the colors that follow - had many subtle
variations and shades. Commonly used descriptive names for
shade and color variations include blue aqua, green or greenish aqua,
pale blue aqua, and so forth. The "gothic" or "cathedral"
style pickle bottle
(ca. 1870-1880) pictured to the left would be described as deep greenish
aqua as it has a distinct light green color to the glass.
(Shades of aqua are a very common color for these utilitarian food
bottles which were a common stylistic design particularly between the
1850s and 1880s though some examples date before or after that period.) The term
aqua is a preferred (by this website) shorthand version of the term
aquamarine. Use of modifying terminology is frequently employed to make more precise
the color shape, intensity, or hue. For example, the fruit
jar pictured below would be considered deep blue aqua.
Aqua glass is a "natural" result of
the iron impurities found in most sands. It is very rare (maybe
unknown) that sand does not contain some traces of iron. Sand deposits with very
low iron content were (and probably still are) highly valued commodities.
Although good quality sand was plentiful in the Eastern United States, some
was still being imported from Belgium for Western American glass factories as
late as the 1940s. Aqua glass is the result of sand which is
relatively low in the amount of iron which was not off-set by
de-colorizing agents as noted in the colorless glass discussion
above. High levels of iron produce darker greens, black glass, and
even amber. Natural aqua glass was often called "green glass," "bottle glass,"
or "bottle glass green"
by glass makers (Kendrick 1968; White 1978).
Different shades of aqua and the
related blue-green colors which are often observed in the same bottles
blown in the same mold may be explained - at least in part - by the
following information quoted from Julian Toulouse's excellent book A
Collector's Manual - Fruit Jars (1969a). This excerpt is making
reference to some of the effects - desired or not - that occur when mixing and
melting glass. It also points out one of the many complexities
inherent in
producing desired glass colors:
Little was known about the
influence of the flame. A "reducing" flame, or one with less oxygen
supplied for burning, might produce a bluish-green because the iron in
the sand might then be reduced to one of the bluer iron oxides - an
excess of air might make the oxidized green iron oxides predominate.
Early glassmakers knew little about this. Thus a fire banked for the
night and with the air intake flues closed down, could produce quite
blue glass for the morning's start, and change slowly during the day
when the air vents were opened wide for a hotter flame.
Diagnostic Utility: Aqua is a very common color in all
types of American made bottles that date prior to the 1920s back at least
to the early 19th century. Thus, no specific type or
class of bottles is more or less apt to have been made with aqua glass.
Aqua
bottles became uncommon after the 1920s when colorless glass
largely replaced aqua as the color of choice for bottle users wishing
their product to be visible to the buyer (Miller & McNichol 2002). The one significant exception to this dating is soda bottles, e.g.,
the greenish aqua of Coca-Cola bottles. Shades of aqua color survived
in soda bottles long after the
1920s and is still found today where glass is still utilized for
soda containers.
Another notable exception is that
many fruit jars were also made well into the the 1930s in aqua though
colorless glass probably became the majority color by about 1930.
Ball fruit jars were made in a distinct "Ball blue" from at least 1909
until at least the late 1930s and even later for some specialty items
(Creswick 1987)). The picture to the left shows two sizes of
Ball's very popular Perfect Mason fruit jar. This color
is light but a more intense blue than blue aqua but does not quite fit the
other color groups described below; thus its coverage here. In 1939,
the Ball Company was making 54.5 percent of all the fruit jars in the U.S.
and the Perfect Mason was one of their most popular lines
(Birmingham 1980). This market domination during the first half of
the 20th century explains the commonness of these fruit jars today and in
historical sites from the early 1900s through the Great Depression.
Opaque white glass - commonly
called milk glass but sometimes called opal or white glass - was typically produced by the addition
of tin or zinc oxide, fluorides (fluorspar), and phosphates (Illinois
Glass Co. 1924; Dillon 1958; Kendrick 1968). It was also created by adding
calcium and phosphate rich animal horns, bones, and even "bat guano" to the glass batch (Trowbridge
1870; McKearin
& Wilson 1978; Jones & Sullivan 1989).
In a sense, milk glass (the preferred term on this website) is like
colorless glass in that it is defined by the absence of color, except in
this case the bottle is truly not "clear". An interesting
feature of most milk glass is that very thin glass (i.e., fragment edge)
has an orange-ish opalescence when held up to bright light. It was
also reputedly hard on the longevity of glass melting tanks and pots so
was avoided by some glass factories.
Diagnostic Utility: Milk glass was
used in the production of a wide array of different type bottles, though
there is some typing and dating utility to the color. The
color was most
commonly used in cosmetic and toiletry bottles (primarily from the 1870s
to about 1920) and ointment/cream jars (1890s to the mid-20th century) (U. of U. 1992). A typical cosmetic
bottle example is the Owl Drug Company lotion (and likely other
cosmetic products) bottle to the left which dates from the around
1895-1910 (Jensen 1967; Fike 1987). Click
milk glass Owl Drug with label to see an example with the original
label showing that particular bottle contained "rose water" which was
a distillate of rose petals used which was a type of natural perfume. Many of the milk glass druggist
type bottles were most likely for that particular druggist's cosmetic
products as milk glass was well linked with cosmetics in customers minds
(Fike 1987; Cannon 2004).
Milk glass was occasionally,
though uncommonly, used for ink, bitters, non-cosmetic medicinals, liquor,
and sometimes even fruit and food
jars primarily during the late 19th and early 20th century. Milk glass is virtually unknown in
the production of soda, mineral water,
wine/champagne, and beer bottles. It was rarely used for bottles
prior to about 1870, i.e., milk glass bottles/fragments would not commonly
be found on historic
site's that date totally prior to the 1870s.
An interesting aspect of some milk
glass is that it was made with manganese dioxide
in the glass batch. This milk glass will react to sunlight - as
described in the colorless glass section above - producing a milky lavender color. This has
primarily been observed in cold cream and other cosmetic jars from the
first couple decades of the 20th century, though it is seen in occasional
bottles. It sometimes appears that the
lavender color was produced purposefully (with a UV light or radiation) so it is not always possible to
differentiate between sun colored and intentionally colored lavender milk
glass.
There
are probably more different shades of green to be found in bottles than
any other color. It rivals the multitude of amber glass variations which, as
noted below, can grade into various greens. The different greens were formed by a myriad of different coloring agents, impurities, and glass making
processes. Iron, chromium, and copper all produce different green
glass. Chromium oxide will produce yellowish green under oxidizing
conditions and emerald green under reducing conditions in the glass
furnace (Dillon 1958). Combinations such as cobalt (blue) mixed with chromium
(green) will, not surprisingly, produce blue-green glass (Kendrick 1968;
Munsey 1970).
Just as there were many ways to
produce different green glasses, there are endless naming variations for
the green colors, e.g., blue-green, clear green, peacock green, jade green, apple-green, emerald green,
grass green, citron, etc. As in
describing all bottle colors, modifiers are useful in clarifying the specific
greens. For example, the flask to the
left would be considered a medium blue-green
with a slight slant towards the green end. This early American (New England)
flask has an eagle motif on one side, a Masonic emblem on the reverse,
and was produced ca. 1815-1825 (McKearin & Wilson 1978).
The mineral water to the right could be described as medium to dark emerald green.
This color is distinctive to mineral water bottles blown at the
Congressville (NY) Glassworks and it and some subtle variations
are known to collectors as "Congressville"
green. This particular Congress & Empire Springs bottle dates
from about 1880-1885 as it has a true
applied finish but also an earlier
mold venting mark on
the
shoulder on each side of the bottle. This distinctive style of bottle is often
referred to as a "Saratoga" style (Tucker 1986). See the
Bottle Typing/Diagnostic Shapes page.
"7-up ®"
green is a intensely brilliant green or yellow-green color that is
associated with the famous soft drink bottle; their plastic bottles
still use the color today. The jar to the left is
a medium density "7-up
®" green and was produced in 1940 based on the information
derived from the Owens-Illinois Co. base markings (Toulouse 1971).
The soda/mineral water bottle to the
right is in a color that could be described as a deep blue-green
with a slant towards the blue end. Compare this color to the medium
blue-green (with a slant towards green) flask in the upper left
corner of this box. This bottle is from the Cottle, Post & Co. (Portland,
OR.) and dates between 1878 and 1880 (Fowler 1975). This bottle was
produced during the era when applied finishes dominated, though it
has an usual - for the time -
tooled finish. It was,
however, not blown in an obviously air vented mold like the majority of
molded & tooled finish bottles.
The flask below has a color that could
be described as clear yellow-green or possibly light
to medium emerald green. This quart flask is a
figured or pictorial Washington-Taylor flask (Philadelphia, PA.)
which was likely blown during the 1860s (McKearin & Wilson 1978).
The bottle to the lower right is an
example of what some call citron after the color of the
fruit of that name. This color is
generally a "brilliant greenish yellow" like the pictured bottle though
citron
is sometimes described as a "brilliant...olive amber with yellow tone"
(McKearin & Wilson 1978). Another description would be "pale
yellowish green with a slight golden cast"
(Spurgeon 2004). The most commonly accepted definition of
citron would be the first description and more or less, the color shown below.
This bottle is a Clement's Tonic (American or Australian) that
likely dates between 1885 and 1900. Citron is somewhat of a color
bridge between the greens described here and the olive greens described
next.
Diagnostic
Utility: Different
colors and shades of green can be found in just about any type or age of
bottle providing no easily described diagnostic utility to
this category of colors. Some limited observations can be made
however:
-The very bright "7-up
®" green is almost
exclusively a 20th century feature as this bright of color is seen on very
few
19th century bottles and then primarily the very late 19th century. The
mouth-blown "7-up
®"
green prescription drug bottle in the middle of the bottle group picture at the top of this
Bottle Colors page is a late handmade bottle dating from about 1915-1920.
-The "Congressville" green illustrated by the Congress Water in the upper
right corner is a color that is most often seen on "Saratoga" style mineral water
(and some soda water) bottles -
primarily from the Eastern seaboard (Tucker 1986). It is a color that was little
if any used in the 20th century and only occasionally shows up in other
types (non mineral water) of bottles - and then primarily 19th century items.
-Similarly, the blue-green colors shown above are very uncommon on
machine-made bottles and generally denote a 19th or very early 20th century manufacture.
These
green colors are distinctly different enough from the greens described
above to address separately. There is also some dating and typing
utility associated with the olive hues which is covered at the bottom. The same
green glass coloring
agents were generally used to produce these olive
colors; see information above. Usually the olive greens and the related green
colors were unavoidable or "natural" colors induced in the glass batch
by variable levels of iron oxide naturally found in the sand. These olive
colors vary widely and grade into the other greens noted
above as well as the amber colors which are discussed next.
There
is also a myriad of descriptive names attached to the range of colors that
have olive green as a significant component: yellow olive,
olive yellow, forest green, olive
emerald, pea green, etc. are
just some of the variations from the basic olive hue As with other
colors, the general degree of density and color can also be described with
modifier terms like deep, dark, medium, light, bright, etc..
The unusual shaped bottle to the
left in a
brilliant olive green is likely a type of sauce bottle
dating from around 1860 as the base has a glass-tipped pontil scar.
This color could be called olive-emerald green because of
the inherent "brilliance" of the glass color.
The decorative or figured flask to
the right is a medium olive amber in color. This small
(1/2 pint) New England made (Keene, NH.)
sunburst flask was manufactured between 1822 and 1830 and
is classified as GVIII-10 by McKearin & Wilson (1978). This
typifies an average olive amber color quite well as it is a fairly
balanced proportion of both colors.
The flask to the left exhibits a
rich bright greenish olive color which is often called forest green.
Forest green is a color variation that seems to be most commonly found in
early (1800-1850) American made bottles and flasks with a color that is
shifted more towards the green than the olive. The pictured
bottle is a "Pitkin" flask manufactured between 1790 and 1830
(McKearin & Wilson 1978). The term "Pitkin" comes from the Pitkin,
Connecticut glass works which is believed to have produced many of these distinctive flasks.
However, Pitkin style flasks were produced in a
pattern mold
using the "German half-post"
method by many different glass makers in New
England and the upper Mid-West during the early 19th century. The
name Pitkin has stuck however as the generic name for all of these flasks
(Buckley 1985).
The
bottle to the right could either be called a moderate
olive green or possibly yellowish olive green,
as there is a slight leaning towards a yellowish tone.
This small, relatively wide mouth bottle is an early American free-blown utility
vessel
which could have been used for
anything from snuff to mustard to medicinals. It likely dates between 1790
and 1820 (McKearin & Wilson 1978).
The
small multi-sided bottle below left is a deep or dark olive amber with
more amber than olive in the color. It is an example showing
where the olive color begins shifting towards amber colors. The
bottle is an eight-sided "umbrella" ink bottle of early (1830s
to 1850s) New
England manufacturing heritage. The color could
also be referred to as dark amber with an olive tone.
Diagnostic
Utility: Different
colors and shades of olive greens and olive ambers can be found in a lot of
different types of bottles from different eras. Generally speaking,
olive green and particularly olive amber are colors which were
much more commonly used in the 19th century than later the 20th - with
some exception noted below. It is also
associated with certain types of bottles much more than other types.
Thus, there is some
diagnostic bottle typing and aging utility with these colors based on
empirical observations. These of course are trends not absolutes, of
which there are few in the world of historic bottle identification:
-Olive amber is a very uncommon color in any type bottle after about 1890
and almost unknown after about 1900. After 1900 it is primarily
found in wine/champagne
bottles.
-Olive greens are very uncommon after about 1900 in most all types of
American made bottles except
some liquor bottles (e.g., scotch) and wine and champagne bottles which
still are made in olive colors. One observation is that the newer
(after about 1920 or so) wine & champagne bottles tend to have a "brighter"
olive green than the more subdued olive green prior to that time.
This color difference would only be evident with much experience looking at such bottles
and is not absolute.
-The olive colors are most commonly found in the following type (age) of
bottles: wine/champagne (any age), mineral water (1880s and before -
particularly in the
"Saratoga" types), ink
(1880s and before), snuff (before about 1870), medicinals (1860s and before),
beer & ale and all types of liquor (prior to 1910),
figured flasks (prior to 1870), and some foods (1880s and before).
-Olive green and olive amber are uncommon colors for
soda water bottles after about 1870 (not real common before that), canning
jars, medicine bottles after the
1860s, perfume/cologne and druggist bottles at any time, and poison bottles. (Covill 1971; Wilson
1972; McKearin & Wilson 1978; Zumwalt 1980; Tucker 1986; Odell 2000)
Various shades of amber in bottles
was very common during the entire age span covered by this website - 19th
through mid-20th centuries. Like the greens, amber colors were
produced from the natural impurities in glass (i.e., iron & manganese) as
well as from color additives such as nickel, sulfur, and in particular carbon, which was added to the glass batch in the form of coal,
charcoal, or even wood chips (Tooley 1953; Dillon 1958; Kendrick 1968).
As
with the other major colors noted, there are many amber variations and
names for those variations. Commonly used terms include yellow,
yellow amber, golden amber,
red or reddish amber, honey amber,
"old" amber (yellowish amber with a distinct greenish
tint), olive amber (covered above), etc. As with the
other colors, the
general degree of density and color can also be described with modifier
terms like deep, dark, medium, light, bright, etc.
The quart beer bottle to the
left is a typical medium amber color - not too dark and not
too light. It is a
Buffalo Brewing Company (Sacramento, CA.) bottle that dates between
1890 and 1902. The precise dating of this bottle was discussed
on the Finish Types & Styles page - click
Finish Types & Styles to review that discussion.
The unusual bottle with an applied handle
to the right has a color best described as reddish amber;
the reddish tint is subtle but distinct in the picture. It is a Wharton's Whiskey (Philadelphia, PA.) that was was blown at the Whitney
Glass Works (Glassboro, NJ) around 1860 (McKearin & Wilson 1978).
Applied glass handles were an unusual feature added to some fancy liquor
bottles primarily during the mid-19th century (1850-1870), though a few
were made to at least as late as the 1890s (Wilson & Wilson 1968).
The tall liquor bottle ("fifth"
size) to the left is an example of what collectors call "old" amber.
More precisely this could be called a medium yellowish amber with an olive
tint, though that makes the name quite a mouthful.
This
particular "4-piece mold" whiskey/liquor bottle was most likely blown at the
San Francisco & Pacific Glass Works (San Francisco, CA.) about 1880 based
on diagnostic features (i.e., crudely applied finish, lack of air
venting, color). The bottle also has a distinctive eight pointed
"star" (like an asterisk) on the base which has been widely attributed
as a product of this Western glassworks (Zumwalt 1980). Click
SF&PGW star mark to view a picture of the base.
The quart fruit jar to the left is
a very light yellow amber or just yellow,
depending on ones interpretation of colors. This is a mouth-blown
Globe fruit jar which were first patented in May of 1886 and
manufactured from that date probably into the early 1900s (Toulouse 1969a;
Leybourne 2002).
Diagnostic
Utility: Since
amber colors were used for an extensive time period in
most types of bottles, there is little dating or typing utility to the
color. However, a few thoughts on amber follow:
-Like olive amber discussed earlier, "old" amber is a color that is very
uncommon in bottles made after 1890.
-In a related vein, when machine-made bottles dominated in the 1920s and
glass making
chemistry and techniques refined, amber became of a
fairly standard tone and density. That is, the off color ambers like
yellow, yellow amber, and black amber (so dark that light hardly
penetrates) become increasingly uncommon and a medium to medium dark amber
-
like the bottle pictured to the left - became the dominate amber shade.
The pictured machine-made beer bottle was produced by the American Bottle
Company (Chicago, IL.) in the 1910s (Lockhart 2004).
-Amber was and still is the most common color for beer bottles as it
provides the best light protection from the light wavelengths that are
responsible for most photochemical reactions. Too much light
(natural or artificial) causes beer to become "light struck" which is a
distinctive "skunky" off-flavor which is associated with green and clear
bottled beer (Papazian 1991,1994). Both of these latter colors
provide little protection from the particular part of the spectrum which
induce the off-flavor chemical reactions whereas amber glass provides a
high level of protection (Dillon 1958).
These
variably moderate to intense blue colors are usually produced with the
addition of the strong coloring agent cobalt oxide to the glass batch.
Copper could also produce types of blue glass depending on the batch
ingredients and melting pot environment (Tooley 1953).
The various blue color shades and
densities give rise to an assortment of names, with cobalt blue
and sapphire blue being the most common covering the darker
and lighter ends of the spectrum, respectively. Dark sapphire
is often used to describe medium cobalt blue; light
cobalt blue is used to describe dark sapphire blue
(McKearin & Wilson 1978). Other descriptive names include:
midnight blue (cobalt so dark it appears black without strong back
lighting), cornflower blue (pale or very light sapphire),
electric blue (a deep brilliant cobalt), and others. However, the usual
modifiers with the two main terms (cobalt and sapphire) are sufficient to
describe most hues and densities of the "true" blue colors.
Blue-green colors were described under the "green" color section above.
"Ball blue" - the very distinctive light greenish blue color
of 20th century Ball® fruit jars - was discussed in the "aqua"
section.
The
bottle pictured to the left above would be referred to as a medium
cobalt blue. It is an Owl Drug Company bottle (San
Francisco, CA. & eventually nationwide) that was produced by the
Whitall, Tatum & Company (Millville, NJ) between about 1895 and 1915.
These bottles reportedly held granulated citrate
of magnesia and/or other products where a wider than average mouth
(bore) was useful in dispensing the product. One example observed by
the author had a label indicating that that at least some (all?)
specifically held "Granular Effervescent Phosphate of Soda" which was a
laxative (empirical observations).
The medium sapphire blue
bottle to the right is embossed on the front with Crystal / Soda
/ Water Co. which was in San Francisco, CA.
These were almost certainly made at the San Francisco & Pacific Glass Works and date
from between 1873 and 1886. The applied finish on this
bottle is somewhat of a modified
blob with a slightly wider ring near the upper end of the lip. This unusual bottle
also has a pedestal base. The actual closure for this bottle was likely a cork and
the ring finish a stylistic feature like the base, though this is not known
for sure. Some versions of this bottle (but not this example) have a
hole on opposing sides of the finish in which a metal pin was pushed through to secure the
cork (Markota 1994). The reverse side notes that the bottle and/or
closure was a result of "Taylor's U.S. PT." (patent) which was "Patented
Nov. 12, 1872." Click
Crystal Soda Water reverse for a picture of the reverse side of the bottle showing this
embossing.
Diagnostic Utility: The presence of the
various noted blue colors has limited dating and typing utility because of
the wide application of the colors for various products. Though not a common color
when compared to aqua, amber, and
the greens, cobalt and sapphire blue can be found to some degree in
virtually any type of bottle from inks to figured flasks to beer bottles
to even occasional food bottles (Covill 1971; Martin 1973; McKearin &
Wilson 1978; Zumwalt 1980). These blue colors are somewhat more
common in certain classes of bottles like those intended for poisonous
substances and cosmetics. An example of the former would be the relatively common early
20th century Owl Drug Company poisons - click
Owl poison bottle for a picture of an Owl poison. These blue colors were also
frequently used with soda and mineral water bottles from the 1840s into
the early 1900s and ink bottles from from the 1840s into at least the
1930s (Schmeiser 1970; Covill 1971; Markota 1994).
Purple,
amethyst and red are uncommon colors in bottles but show up with
enough frequency to warrant mention. This group of purple to reddish
colors were usually a result of glass that was colorized with nickel or
manganese oxides (Tooley 1953) with true red usually a result of the use
of oxide of gold (Hunter 1950). As noted in the colorless
glass description above, small amounts of manganese dioxide was used as a
decolorizer to offset the iron impurities present in virtually all sands.
This colorless glass will variably turn amethyst upon long
term exposure to sunlight. With larger concentrations of these
substances in the
glass batch amethyst to purple glass is purposefully created
(Jones & Sullivan 1989).
Because of this
color's variability (and popularity with collectors) it is not surprising that there are numerous names for
subtle differences in this
color theme. They include descriptive words based on real reddish
substances like claret, burgundy, red
wine, or if tending towards amber, puce which
according to Webster's Unabridged Dictionary (1996) is a "dark purplish
brown to dark red." The difference between purple and amethyst
is subtle, though amethyst is often a "pinker" color than purple.
True red colored utilitarian bottles are very unusual and usually referred
to as ruby red.
Diagnostic Utility: These
true purple/reddish/amethyst colors (not sun colored amethyst) in bottles are
primarily found in the era between the 1840s and early 1880s; they are
rarely noted in bottles that date before or after that date range
(empirical observations).
The deep reddish amethyst
colored bottle above left is a Mrs. S. A. Allen's World's Hair Restorer
(New York) which dates from the 1870s. Hair treatment bottles are
one of the few classes of bottles in which the purple/amethyst colors are
fairly commonly found. Other classes of bottles where these colors
occur with some frequency (though still not commonly) is figured or
pictorial flasks, bitters (particular those which are "barrel" shaped
like the bottle to the right),
and some types of ink bottles. The bottle to the right is
an Old Sachem Bitters and Wigwam Tonic (New York) which likely
dates 1860-1870. It's color would be described as medium amethyst.
(This photo courtesy of Jeff Noordsy Antiques.)
True red - ruby red - utilitarian bottles are only known to the author
as having been used for bottling Schlitz™
beer at various times - and in various period beer shapes - from the late
1940s to early 1960s (empirical observations). An image of a
Schlitz™ quart
stubbie style beer bottle is to the left. Click
base view to see such showing the makers mark for the Anchor
Hocking Glass Co. who made all of these esthetically colored bottles
for Schlitz™. The base photo also
shows the "49" date code indicating production in 1949. (Photos from
eBay®.)
Black glass is
probably "...the most important of the green glasses..."
which "...was of so deep a color as to appear black in reflected light
and even in direct light when the walls of the bottles were very thick..." according to McKearin & Wilson (1978:9). This strong and resilient
glass was also a color that offered the
most protection to the contents from the effects of direct light.
Most black
glass bottles are actually a very dark olive green or olive amber.
These types of black glass were the result of the same impurities or
coloring agents as the olive colors - most usually high iron
concentrations but also other substances including carbon (from various
sources including ashes and coal clinkers), copper with iron, and magnesia
(Jones & Sullivan 1989; McKearin & Wilson 1978). Occasionally, black glass
can be very dark amber ("black amber" - example shown later) or very dark reddish purple
("black amethyst"), although these types
are not remotely as common as the olive color shades. When the
term "black glass" is used on this site, it is making reference to the very dark olive
colors unless noted.
Black
glass is one of the oldest bottle colors going back historically well
beyond the time frames covered by this website - to at least the mid-17th
century in Europe (Van den Bossche 2001). Black glass
beverage bottles (known as "junk bottles" by early glassmakers) and
fragments are ubiquitous on historic sites that date prior to 1880
(McKearin & Wilson 1978:229-232). Black glass
liquor and ale/beer bottles "...were mass produced as a cheap container
between the 1840s and 1880s...in a thousand shapes and sizes" (Wilson
& Wilson 1968). There is no measurable line delineating the point
where olive green and olive amber become black glass except as noted at
the beginning of this section as glass that appears "...black in
reflected light..." (McKearin & Wilson 1978).
The accurate dating of
late 18th and 19th century
black glass containers is difficult for several reasons, most notably for
the reason summarized by the following - "As late as 1880 the San
Francisco glass houses were turning out (black glass) bottles as
crude as those made in the east many years earlier, mainly because of
unskilled labor and the inability to retain the workers" (Wilson &
Wilson 1968). In mid-19th century frontier California, where
transportation costs from the east coast were prohibitive, the local glass
makers enjoyed somewhat of a geographic monopoly. This protective
situation provided little incentive for glass makers to quickly adapt new
methods for the production of cheap utilitarian wares such as black glass
liquor and ale bottles. This was not the situation on the east coast
where glass makers had to fend off cheap imports from Europe, the amount
of
which flowing into the country was dependent on the level of tariffs and duties at any given time
(Davis 1949).
The square, black glass (very dark forest green) bottle to the above left is a Hostetter's Stomach Bitters
(Pittsburgh, PA.) that likely dates from between the late 1850s and
mid-1860s. (Note: this
bottle is also used as a dating example on the
Examples page.) Square bitters, large medicinals and liquor bottles
- similar to the Hostetter's and in black glass - were a fairly
common item made between the late 1840s and about 1870 (Switzer 1974;
Wilson & Wilson 1968 & 1969).
The black glass (very dark olive
amber) liquor or ale bottle to the above right is of early American origin being
blown by the New England Glass Bottle Company (Cambridge, Mass.).
This company operated between 1827 and 1845 (McKearin & Wilson 1978). The company name is
embossed very faintly on the base of this bottle - click
NEGBCo base marking to view a picture of the base. The middle portion of
the base has a sand pontil mark that is typical of that era. This
bottle is of very thick glass so that it is quite black even when backlit.
The very dark olive
or "light" black glass bottle to the left is a snuff bottle
that was most likely made in New England in the 1850s, as it has a glass
tipped pontil scar on the base although it was unearthed in the Pacific
Northwest. This color is transitional between
what is called black glass and very dark olive green though appears black
in reflected light though not with moderate back lighting shown in the
image. These two
colors are usually the same actual glass color with the black glass versions being
either thicker glass or a denser tone.
Diagnostic Utility: American made black glass bottles
of any type were uncommon after about 1880, making the presence of this
color useful in the dating of archaeological sites (Wilson & Wilson 1968;
empirical observations).
Even in imported bottles, black glass seems to disappear during the
1890s and those bottles that were earlier made in black glass are lighter
in color after that time, i.e., medium olive green or olive amber
(empirical observations), though as previously noted, there is no firm
transition point from where the lighter colors end and black glass begins.
The majority of black glass bottles made during
the 19th century were for liquor or for wine and ale for which protection
from the light was important in retaining quality. This color is
also found in pre-1870s ink bottles (ink
bottles, ink wells and bulk inks), mineral waters (particularly the
"Saratoga" types),
snuff bottles, and some
earlier medicinals and rarely for food bottles (empirical
observations).
Most of the dating and typing
comments listed for the olive colors earlier on this page hold also for black glass
except that black glass was little used in the 20th century for
bottled products. This includes wine, champagne, and liquor bottles
which - in the 20th century - could be (if olive colored) of fairly dark
olive green or olive amber glass but which would not meet the "black in
reflected light" definition noted above (Covill
1971; Wilson 1972; McKearin & Wilson 1978; Zumwalt 1980; Tucker 1986;
Odell 2000). One notable exception to the last statement are the black amber soda bottles produced for Mission Dry Orange
(soda) between 1929 and the mid 1930s. The image to the right shows
an example of these bottles which are discussed in more depth on the
Soda & Mineral Water Bottles typology page.
SEARCHING THIS WEBSITE: To do a word/phrase search of this website
one must use the "Search SHA" boxes found on many of the main SHA
web pages,
including the
Research Resources page (upper right side of that page) which links to
this site. The Historic Bottle Website (HBW) has no internal search
mechanism so be aware that when running a search one will also get non-HBW
response links to other portions of the SHA site.